Introduction to Epidemiology:
the basic science of public
health
Al Katz, MD, MPH
John A. Burns School of Medicine
University of Hawaii
How does “public” health differ from an
individual “clinical” approach to health?
• Focus is on populations rather than
individuals
– International
– National
– State
– City
– Defined group
How does “public” health differ from an
individual “clinical” approach to health?
• Focus on prevention rather than cure
– Surveillance
– Immunizations
– Sanitation
– Assuring clean water
• Assessment of the health of population = collection
and analysis of information from the population
• Potential for greater impact:
Impact of public health is limitless
• WHO Smallpox eradication program
– In 1967: endemic in 33 countries with
estimated 10-15 million cases
– Through intensive surveillance, immunization,
and containment: last case reported in 1978
– Officially declared eradicated in 1980
Definition of epidemiology
• The study of the distribution and
determinants of health related events in
human populations
Descriptive epidemiology
• The study of the distribution of health
related events by:
• Person (who)
• Place (where)
• Time (when)
Person
• Age
• Sex
• Ethnicity
• Socioeconomic status
• Marital status
• Occupation
• Behavioral practices
Place
Geographic distribution
• Urban versus rural
• Variation by province
• Clustering of cases within provinces?
Time
• Temporal trends in disease occurrence
• Seasonal fluctuations
• Incubation period: time from exposure until
onset of symptoms
Descriptive epidemiology
• Important initial step in gaining an
understanding of disease occurrence
• Gives clues to possible disease risk
factors
Analytic epidemiology
• The study of the determinants or causes
of disease
• May be observational or experimental
Analytic epidemiology
Experimental Observational
Studies Studies
Clinical Trials Cross-sectional*
Community Cohort
Intervention Trials Case-control
*may be considered either descriptive or weak
analytic
Epidemiologists do “comparison”
studies
• How do events in one population compare with
those in another?
• How do persons of differing genders experience
a disease?
• How do persons of different ages experience a
disease?
• How do the ill differ from the well?
• How do the exposed differ from the unexposed?
Comparison studies focus on rates
• Need to measure and compare rates of
disease occurrence between different
populations or population subgroups
Rates take into account both
numerator and denominator data
• Numerator data = number of events
• Denominator data = population at risk for
experiencing the event
Use of rates for comparison purposes
Rate = Number of events in a specified time
Population at risk for the event
Example: for 2005, UN estimated number of deaths in China = 8.8
million, but only 44,000 deaths in Botswana.
Is Botswana a healthier country than China?
Differences in population size need to be taken into account
before one can start to compare the death rates in these two
countries
Rates used by epidemiologists
• Mortality rates: rates of death
• May be crude, or made specific to different
factors such as age, sex, or disease
Rates used by epidemiologists
• Morbidity rates: Rates of disease
occurrence
– Incidence rates: rates of new onset disease =
risk
– Prevalence (rates) proportion: measurement
of existing disease (both old and new cases)
= burden of disease
Example of incidence and
prevalence
Prevalence of HIV infection among gay
men in San Francisco California is
approximately 25%; the incidence rate
of HIV infection in this same population
is approximately 1-2%.
Basic tenet of epidemiology
• Disease is not randomly distributed in the
population
• Different population subgroups may
experience disease occurrence differently:
by exploring why, clues to disease
causation or prevention may be found
Injuries are not accidents: focus on
motor vehicle injuries
Person factors:
• New drivers, especially teens, are at high
risk for motor vehicle injuries
• Alcohol consumption and speed are risk
factors
• Not using seatbelts
Injuries are not accidents
Environmental factors:
• Poor lighting
• Improperly designed or maintained
roadways
• Fixed objects near roadway
Uses of epidemiology-1
• Thorough understanding is needed to
critically review the medical and public
health literature:
– Adequate sample size?
– Bias in the study design?
– Use of appropriate statistical tests?
Critically reviewing the medical and
public health literature
• It is not enough to read the abstract and
the discussion
• Need to critically review the “Materials and
Methods” section
Uses of epidemiology-2
• Necessary to interpret “screening tests”
– Even with high sensitivity and specificity,
positive screening test results may be false if
the prevalence of the disease or trait is low in
the population being screened
Interpretation of screening test
results
• A PCR test for gonorrhea (Roche
amplicor) has a sensitivity of 98.2% and a
specificity of 99.3%
• If this test is used to test for gonorrhea in a
population where there is a 1% prevalence
of gonorrhea, what proportion of persons
testing positive will be mislabled (false
positives)?
Interpretation of screening tests-II
• Answer: the positive predictive value of
this test when applied to this population is
60%
• This means that when this test is applied
to this population, 40% of those screening
positive will be false positives!
Uses of epidemiology-3
• Design and implementation of sound studies to
investigate health related events
• Well designed epidemiologic studies have been
invaluable to disease understanding and control:
Examples:
– Demonstration that fecally contaminated water is a risk
factor for cholera (John Snow)
– Handwashing can prevent nosocomial infections
(Semmelweis)
– Tobacco is a risk factor for lung cancer (Doll & Hill)
Need for epidemiology to develop
and interpret studies
• Poorly designed studies have caused
harm. Examples:
– Purported protective effect of DES on adverse
pregnancy outcomes
– Purported widescale, generalized risk of
autism associated with routine childhood
immunization
Epidemiology helps the clinician
• Epidemiologic information assists the
clinician in making an accurate diagnosis
and prescribing the appropriate
medications
• Examples:
– Correct choice of antibiotic regimen based on
surveillance of drug sensitivity patterns of
bacterial isolates
– Correctly diagnosing symptom complexes
based on common epidemiologic patterns
Epidemiology is important for
health care planning
• Disease surveillance provides information
on trends and allows health planners
to allocate resources more responsibly and
cost-efficiently.
–Infectious diseases: vaccine preventable
illnesses; foodborne illnesses, STDs, HIV
–Chronic diseases: cardiovascular diseases,
cancer
• Helps lay the groundwork for prevention
and control programs
Prevention Levels in Public Health
• Primary: Prevent the cause of the disease
• Secondary: Early detection and prompt
treatment of the disease in its early stages
• Tertiary : Limitation of disability and
assisting with rehabilitation where disease
has already occurred and left residual
damage
Primary Prevention
• Immunizations
• High fruit and vegetable, low saturated fat
diet
• Regular exercise
• Condoms to prevent sexually transmitted
diseases
Secondary Prevention
• Routine annual mammography for women
starting at age 40
• Colorectal cancer screening for persons age
50 and older
• Chlamydia screening for sexually active
women < 25
• Routine Pap smear to identify cervical cancer
and pre-cancerous cervical disease
