Introduction To Clinical Epidemiology

DR.Dr.
JULIANDI HARAHAP, MA
Dept. of Communitiy Medicine
FK USU
 Sciencewhich is about the distribution of
disease (or consequence of disease) in
populations
 Set
of methods used to reduce uncertainty
about information
The study of the distribution and determinants of
disease in human populations.
Study
 Methods are intended to be scientific (basic science of public
health)
 Epidemiology is reasoned argument
Distribution
 Descriptive Epi – person, place, time
 Look for patterns among different groups
Determinants
 Epi Triad – Agent, Host, Environment
 Causality – Criteria (AB Hill), Induction vs. Deduction (Popper)
Populations
 Probability (chance) and Statistics
 Study Designs
 CE is the science of making predictions about
individual patiens by counting clinical events in
groups of similar patients and using strong
scientific methods to ensure that the predictions
are accurate.
 Aims:
1. Provide clinicians with optimal strategies for
diagnosis, treatment and prognosis,
improvement of clinical decision making.
2. Better clinical decision results from studies
evidencing real facts, evidence based
medicine.
Used as an aid to clinical decision making

Clinical Research Epidemiology
Patient-oriented Population-oriented
research research
 context  methods
Clinical
Epidemiology
A basic science for clinical medicine
(Sackett et al, 1991)
1. Formulate question (hypothesis)
2. Choose study design
3. Choose study population base and sample
from that base
4. Collect and analyze data
5. Interpret results
Elements of a good question (hypothesis)
 Target population
 Target exposure (risk factor, protective factor,
treatment) and its control
 Specified outcome
 Example: In adults is hypertension associated with an
increased risk of coronary artery disease?
Criteria for a good question

 New information?
 Feasible to answer?
 Public health importance?
A. Experimental (e.g. RCT)
B. Observational
1) Cohort (incidence) study
2) Case-control study
3) Cross-sectional (survey, prevalence) study
 Also called randomized, controlled trial (RCT)
 Randomly assign “exposure” (usually an
intervention) to study population
 Measure “outcome” during some follow-up period
 Best method to eliminate selection bias (required
for pharmacologic treatment by FDA)
However
 – Some exposures cannot be randomly allocated
 – Selection criteria may limit generalizability
 – High cost
 Classify individuals in study population into
“exposure” categories
 Measure “outcome” during some follow-up period
 Can be retrospective or prospective
 Feasible for common outcomes, less costly than
RCTs
However
 – Subject to selection bias
 – Among observational study designs most costly
 – Not feasible for rare outcomes
 Choose a sample of individuals with the
“outcome” of interest (cases) and a sample of
individuals without the “outcome” (controls)
 Assess “exposure” status of study population
sometimes in the past
 Retrospective except for nested case-control
studies, i.e. when case and control subjects are
drawn from a cohort study
 Advantages: less costly, appropriate for rare
outcomes
 Disadvantages: more subject to bias
 Also called a prevalence study
 Assess “exposure” and “outcome” of study
population at the same time
 Advantage: least expensive of observational
studies
 Disadvantage: no temporal relationship between
“exposure” and “outcome”
 Struggle between access and generalizability
 Population Base choices include
– Active patients (clinic/hospital)
– Community residents (geographic)
– Combination
 Sampling choices include
– Random sampling
– Stratified random sampling
– Convenience sampling
 Depends on study design
 Sensitivity and specificity in studies of
diagnostic tests
 Relative risk and odds ratios in studies of
association between risk factors and
outcomes
 Threats to the validity of epidemiologic
studies
- Bias (a process at any stage of

inference tending to produce results that
depart systematically from the true
values)
- Chance
Internal validity
 The degree to which the results of a study
are correct for the sample of patients being
studied
External validity
 The degree to which the results of an
observation hold true in other settings
Selection bias
 Occurs when selection and/or follow-up procedures
lead to study group differences in determinants of
outcome other than the one under study
Measurement bias
 Occurs when measurements are imprecise and/or the
methods of measurement are dissimilar among study
groups
Confounding bias
 Occurs when two factors are associated (“travel
together”) and the effect of one is confused with or
distorted by the effect of another)
 Due to selection or by chance
 Restriction (exclude the young)
 Matching (for every 50 year old with BP+,
find a 50 year old with BP-
 Stratification at enrollment (set sample size
for each decade of age)
 Statistical adjustment (stratification of data,
standardization, multivariate adjustment)
 Use “objective” and precise measures when
possible
 Blinding of assessors of outcome to risk
status of subject and vice versa
 Assess extent of measurement (information)
bias by comparing responses to external
information (e.g. pharmacy records, or
health utilization data bases)
 Arises from sampling and measurement
 Assessed by statistical methods

Introduction To Clinical Epidemiology

Uploaded by

Copyright:

Available Formats

Introduction To Clinical Epidemiology

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Introduction To Clinical Epidemiology

Uploaded by

Copyright:

Available Formats

DR.Dr.

Used as an aid to clinical decision making

Criteria for a good question

- Bias (a process at any stage of

You might also like