INTRODUCTION TO EPIDEMIOLOGY AND BIOSTATISTICS

1
INTRODUCTION TO EPIDEMIOLOGY AND
BIOSTATISTICS Spring 2005
2
INTRODUCTION TO EPIDEMIOLOGY AND
BIOSTATISTICS Spring 2005
Lecture 1 Overview and Causation in
Epidemiologic Investigation Lecturer Hoda
Anton-Culver, PhD Professor and Chief of
Epidemiology Department of Medicine Date
Time April 8th, 100 250pm Topics Ov
erview of Epidemiology and Current
Trends Association versus causation, criteria
for asserting cause Students are responsible
for the book assignment Assignment Read
Jekel Chapters 4, pp 65-70
3
BROAD DEFINITION OF EPIDEMIOLOGY

• The study of the distribution and determinants of
  health-related states or events in specified
  populations and the application of this study to
  the control of health problems.
• (Last, JM. Dictionary of Epidemiology)

4
PURPOSE OF EPIDEMIOLOGY

• To describe the frequency and extent of health
  conditions and disease.
• To determine the burden of disease in a
  community, including socioeconomic impact of
  disease occurrence in specific populations.
• To identify the causes and risk factors of
  specific diseases. This is the basis of disease
  prevention.
• To evaluate medical interventions, including both
  preventive and therapeutic measures, and evaluate
  the delivery of these measures in health care
  settings.
• To study the natural history and prognosis of
  disease.
• To provide the foundation for developing public
  policy and regulatory decisions relating to
  health.

5
EPIDEMIOLOGIC DESCRIPTION OF DISEASE

• Characteristics of persons
• a. Age
• b. Sex
• c. Ethnic groups
• d. Religion
• e. Marital status
• f. Occupation
• g. Socioeconomic Status
• Place
• a. Types of comparisons
• International
• Rural Urban
• Local distributions

6
EPIDEMIOLOGIC DESCRIPTION OF DISEASE

• Place (cont)
• b. Characteristics peculiar to place
• 1. Biologic environment
• 2. Chemical and physical environment
• 3. Social environment
• Time
• a. Point epidemics
• b. Secular change
• c. Cyclic fluctuations
• d. Clustering in time
• Combination of person, place and time
• a. Migrant studies
• b. Birth cohort studies
• c. Time place clustering

7
Schematic representation of the natural history
of disease

• STAGE OF SUSCEPTIBILTY PRE- CLINICAL DIS
  ABILITY
• DISEASE SYMPTO- DISEASE OR RECOVERY
• MATIC
• TISSUE Resolution or
• CHANGES Pre-Pathogenesis Pathogenesis
  Sequelae
• LEVEL OF
• PREVENTION Primary Secondary Tertiary
• MODES OF Health Promotion Detection Treatment
  and
• INTERVENTION Specific Early Rehabilitation
  
• Protection Diagnosis Limitation of
• Prompt Disability

Outcome Death
E
Clinical Disease
Chronicity
B
D
C
A
Clinical Horizon
Residual Disability
Inapparent or Subclinical
8
Idealized Conceptualization of the Scientific
Method
Theory/ Knowledge/ Problems
Synthesis Hunches
Inferences Re Conceptual Ho
Conclusions and Interpretations
Conceptual Hypotheses
Inferences Re Operational Ho
Study Design
Operational Hypotheses
Empirical Findings
Data Collection
Observations/ Data
Data Analysis
9

• Diversity in Approach of Epidemiologic
• Assessment of the Distribution and Cause of
  Disease in Human Populations
• Population Research
• Healthcare aspects Research
• Prevention Research
• Basic mechanisms using multiple systems
• (cell cultures, stem celletc)
• Genetic/ family-based research
• Childhood disease research
• Race/Ethnicity issues
• Aging and geriatric research
• Psychosocial/ quality of life research
• Clinical and Therapeutic trials

10
Health Care Outcome Approachs
Management
Outcome Measures
Patient
Health Care Provider

• survival
• recurrence
• new disease
• treatment complications
• health related quality of life
• disease-specific impairments
• satisfaction with care
• financial family burden

• medical specialty
• medical coverage/insurance
• access to clinical trials
• support group availability

• prognostic factors
• knowledge/interest learning about cancer
• coping strategies
• quality of life

Treatment/Diagnostic

• treatment recommendation
• treatment given
• appropriateness of treatment
• pain/symptom management

11
Example of a Pedigree Generated from Expanded
Family History Data File
UC Irvine Epidemiology Division 11/8/2000
Page 11
12
FAMILY HISTORY
Family history is a significant component to
facilitate research studies on a)
Susceptibility to human disease b) Early
detection and prevention c) Genetic testing
including psychosocial/behavioral issues d)
Identification and characterization of
populations at high genetic risk e)
Chemoprevention and other clinical trials f)
Other translational research
13

• Clinical Research Approaches
• The risks are highest
• Professional roles are conflicted
• Conflict in two directions, to do no harm and
  randomization where 50 of the patients will not
  benefit
• Clinical research is a multibillion Dollar
  business and has a high potential for profit
  riding on efficiency aggressiveness and positive
  outcome
• Given the complexity of human research, can
  consent ever be truly informed?

14
Types of Frequency Measures

• Incidence Measures
• The numerator reflects the number of new
  (incident) cases of a disease identified during a
  given period.
• Prevalence Measure
• The numerator reflects the number of existing
  (prevalent) cases of a disease identified at a
  point in time or during a given period.
• Mortality Measures
• The numerator reflects the number of deaths due
  to or with one or more diseases, identified
  during a given period.

15
Incidence Rates

• Incidence rates are the fundamental measure for
  etiologic studies since they are direct
  indicators of risk of disease over a period of
  time.
• High incidence rates are synonymous with high
  risk of disease.
• Measurement of incidence requires at least two
  sets of observations to determine the number of
  new cases of disease that have occurred among the
  initial disease-free individuals during some
  specified period of time.
• By combining information on incidence rates with
  other variables, we can determine what factors
  may affect the risk of acquiring disease.

16
Prevalence Rates

• In contrast to incidence rate, prevalence rate is
  a static concept. It provides us with an
  indication of the amount of disease prevailing in
  a given point in time (in a period), thus allows
  assessment of the burden of a particular disease
  in a community.
• Prevalence depends on
• Number of people who have become ill in the past
  (previous incidence).
• The duration of their illness.
• If 1 and 2 are fairly stable, then
• Prevalence Incidence x Duration

17
The following characteristics should be
remembered concerning incidence and prevalence

• Causes and effect, such as smoking and lung
  cancer, if studied simultaneously, it would be
  impossible to establish a time sequence for a
  presumed cause.
• High prevalence rates may reflect either high
  incidence or long duration.

18
A causal association is one in which a change in
the frequency or quality of an exposure or
characteristic results in a corresponding change
in the frequency of the disease or outcome of
interest.
Causation of Disease
19
Types of Causal Relationships

• Sufficient cause If the factor (cause) is
  present, the effect (disease) will always occur
  e.g. Genetic disorder such as Tay-Sachs disease.
• Necessary cause If the factor (cause) is absent,
  the effect (disease) cannot occur e.g.
  Tuberculosis.
• Risk factor If the factor is present and active,
  the probability that the disease will occur is
  increased e.g. Cigarette Smoking
• Directly causal association The factor exerts
  its effect in the absence of intermediary factor
  (intervening variable) e.g. Accidents
• Indirectly causal association The factor exerts
  its effect via intermediary factor e.g. SES.
• Non-causal association The relationship between
  two variable is statistically significant, but no
  causal relationship exists

20
Guidelines for Disease Causation
21
Common Pitfalls in Causal Research
22
Human Research Approach and Design
Analytic Examine etiology, efficacy
Descriptive To Describe an experience,
programs, treatment, unusual observation
Experimental Evaluate the efficacy of a
therapeutic , or other interventions
Observational Association of cause and
effect Comparison between 2 treatments
Examples 1) Case Reports or series Side
effect of a drug Cluster of cases 2)
Clinical series Cases with West Nile Virus 3)
Population Prevalence or incidence of
disease in populations 4) Ecological study

• Examples
• Clinical Trial
• Community education

Examples 1) Cross-sectional 2) Case-control 3)
Prospective/Cohort
23
Ecological Study Design

• The strategy is based on determining whether
  those ecological units with a high frequency of
  exposure also tend to be the groups with a high
  frequency of health outcome occurrence.
• Measure group rates of the health outcome and
  exposure prevalence for the same population
  group, but not necessarily using the same source
  of data.
• Types of comparisons
• Geographical or group comparisons, e.g.,
  countries or administrative units.
• Temporal comparisons
• Combines geographic and temporal comparisons
• Time-series studies.

24
Cross-Sectional Study

• Cross-sectional studies typically conducting a
  medical survey in a community or defines
  population. Key steps in conducting a
  cross-sectional medical survey are
• To identify the base population
• To choose a sampling design and sampling frame
  for selecting the study participants
• To measure exposure and health outcome status on
  the study subjects.
• A critical aspect of conducting a medical surveys
  is the logistical plans for examining the
  subjects and obtaining and processing any
  biological specimens.
• Cross-sectional studies may also be based on
  existing records. For example, a study used data
  from NHANES to evaluate the association between
  lead exposure and hypertension by comparing blood
  lead concentrations and blood pressure in the
  survey participants (Schwartz, 1986).

25
Cross-Sectional Study
EC

• Syn Prevalence study survey
• A nondirectional or backward design involving
  disease prevalence (C/C) and random sampling from
  a single source population (N), which may be
  stratified. Although cross-sectional studies
  maybe conducted without random sampling, this
  strategy offers little potential for describing a
  larger population or testing etiologic
  hypotheses.
• Cross-sectional studies are often used to study
  conditions that (1). Are measured on a
  continuous scale and can vary over time (2).
  Are relatively frequent diseases with relatively
  long durations of expression (e.g., nonfatal and
  incurable conditions).

_ EC
_ EC
N
S
__ EC
26
Case Control Study

• Is a study that starts with the identification of
  persons with the outcome of interest and a
  suitable control group of persons without the
  outcome. The relationship of an exposure to the
  outcome is examined by comparing those with and
  without the outcome with regard to how frequently
  the exposure is present, or if quantitative, the
  levels of exposure in each of the groups
• Syn Case-referent Study Retrospective Study

27
Case Control Study

• There are two alternative approaches for sampling
  controls in case-control studies involving
  incident cases
• Density Sampling one or more controls are
  selected for each case ay the time of case
  detection i.e., matching on time.
• Cumulative Sampling all controls are selected at
  the end of the observation period during which
  the cases are identified.
• Density sampling is generally preferable when the
  observation period is long, especially if the
  frequency of exposure changes over time

28
Case-Control Study

• A backward or nondirectional design in which a
  group of cases- either incident (D, as above) or
  prevalent- are compared to a group of noncases or
  controls ( C ) with respect to previous or
  current exposure status. The ratio of controls-
  to cases is fixed by the investigator. To make
  causal inferences, the investigator must assume
  that the control group ( C ) is representative of
  the same source population (N) from which the
  cases (D) developed.
• The cases-control study differs from the
  cross-sectional study in two ways In a case
  control study, subject are samples from two
  populations (cases and noncases) and cases may be
  incident or prevalent in a cross-sectional
  study, subjects are sampled from one population
  and cases are prevalent only. (cont.)

DE
_ DE
N
D
_ CE
N
S
_ C
__ CE
29
COHORT STUDIES

• (also called follow-up studies, longitudinal
  studies, and incidence studies)
• Cohort study is the method in which subsets of a
  defined population can be identified who are,
  have been, or in the future may be exposed or not
  exposed to a factor or factors hypothesized to
  influence the probability of occurrence of a
  given disease or other outcome.

30
1. Prospective Cohort Studies. When cohort(s)
are assembled in the present and followed into
the future.

• Specific Exposure Cohorts.
• When the disease outcomes of a cohort with a
  specific exposure are compared to the same
  disease outcomes of an appropriate unexposed
  cohort. Specific exposure cohorts are preferred
  when the exposure is unique or relatively rare
  (e.g. Hiroshima survivors).
• General Population Cohorts
• When the cohort is a sample of the general
  population all of whom are followed for the
  disease outcome. The exposure status of
  individuals members of a general population
  cohort are assessed at the beginning of the study
  (and sometimes periodically throughout the
  study). Cohort studies based on general
  populations are preferred when the exposure of
  interest is relatively common and/or when there
  is interest in investigating more than one
  exposure simultaneously.

31
2. Historical Cohort Studies

• (also called Retrospective Cohort Studies).
• When the cohort(s) are assembled from past
  records. Follow-up can be either in the present
  or from more recent past records. Historical
  cohort studies may study either general
  population cohorts or specific exposure cohorts.

32
The Structure of a Cohort Study
DISEASE
EXPOSED
NON-CASES
NO DISEASE
RANDOM SAMPLING
SOURCE POPULATION
DISEASE
NON-EXPOSED
NO DISEASE
PREVALENT CASES
FOLLOW-UP OUTCOME
33
Figure 5-2 Design of cohort study

• TIME
• Exposure Disease
• Study group Present
• Comparison
• Group Absent
• Disease rate for exposed a
• a b
• for nonexposed c
• c d
• a
• a b
• Relative Risk c
• c d

a
Present
b
Absent
c
Present
Absent
d
34
Sampling strategies for Epidemiologic Studies
Population
Exposed
Random Sample
Controls
Cases
Unexposed
35
STRATEGY FOR REVENTION

• (1) RISK ASSESSMENT
• A) EXPOSURE.
• MONITOR POTENTIAL CARCINOGENS IN AIR, IN
  WATER, SOIL AND THE HOME.
• USE DATA FROM REGISTRIES OF ENVIRONMENTAL
  ACCIDENTS.
• B) POPULATIONS AT HIGH RISK FOR CANCER.
• IDENTIFY AND MONITOR FOR INCIDENCE AND
  PREVALENCE

36
STRATEGY FOR REVENTION

• (2) INTERVENTION
• A) EDUCATION AND HEALTH PROMOTION
• B) CONTROL EXPOSURE TO KNOWN CARCINOGENS
• C) POPULATION TRIALS
• IMMUNIZATION, CHEMO-PREVENTION AND
• BIOLOGICAL MARKERS FOR EARLY DETECTION
  
• D) GENETIC TESTING AND PSYCHOSOCIAL SUPPORT
• E) EARLY DETECTION

37
STRATEGY FOR PREVENTION

• (3) EVALUATION
• A) MEASURE CHANGE IN RISK
• USING CPRR's MONITORING OF INCIDENCE,
  STAGE AT DIAGNOSIS, USE OF HEALTHIER LIFESTYLE
  PREVENTION PRACTICES.
• (4) MODIFY INTERVENTION STRATEGIES
• A) BASED ON EVALUATION RESULTS
• B) BASED ON NEW SCIENTIFIC FINDINGS

38
Strategy for Prevention
Identify
A
Populations
s
s
at High
Modify Existing
e
Disease Risk
s
Intervention
s
(based on demography / family history,
host factors..)
m
Programs

e
n
n
o
t

i

t


n

Assess

e

Evaluate

v
Exposure


r


Intervention
e



t
Programs
n
I
Conduct
Research on

Mechanisms
Apply
(including the study of genetic susceptibility)

Population-Based
Intervention
Programs
Epidemiology Division