Epidemiology and Quantitative Methodspart A

1
Epidemiology and Quantitative Methods-part A

• Kayla Laserson
• TLC Course
• April 2001

2
Epidemiology

• The study of the distribution and determinants of
  disease frequency

3
Epidemiology

• Disease frequency
• Quantification of the existence or occurrence of
  disease (or event)
• Disease distribution
• Who gets the disease? Where is the disease
  occurring? When is the disease occurring?
• May involve comparisons between different
  populations at a given time, between subgroups of
  the population, or between various periods of
  observation
• Disease determinants
• Combination of first two

4
Epidemiologic Reasoning and Process (1)

• First Suspicion concerning the possible
  influence of a particular factor on the
  occurrence of disease/event (e.g., men and women
  have different default rates)
• Suspicion may arise from clinical practice,
  examination of disease patterns, observations, or
  from speculation
• Second Generate hypothesis (e.g., more men than
  women default TB treatment)
• Third Test hypothesis in epidemiological study
  or from collected data that includes appropriate
  comparison group (e.g., examine TB registry)

5
Epidemiologic Reasoning and Process (2)

• Fourth Data collection and analysis
• The systematic collection and analysis of data
  involves the determination of whether a
  statistical association exists
• Is the probability of developing a particular
  outcome (defaulting TB treatment) in the presence
  of a given factor/ exposure (men) different from
  the corresponding probability in its absence
  (among women)?

6
Epidemiologic Reasoning and Process (3)

• Fifth Assess validity of any observed
  association
• Exclude any possible alternative explanations
• Chance
• Systematic errors in data collection or
  interpretation (bias)
• Other variables that might be responsible for the
  observed association confounding (e.g., more
  men have HIV and this leads to default)
• Finally Make judgement if a true cause-effect
  relationship exists between factor and outcome

7
Measures of Disease Frequency (1)

• The measures used most often are prevalence
  incidence
• Prevalence proportion of individuals in a
  population who have the disease (event) at a
  specific instant
• P with disease or event/ total in
  population
• at a given point in time

8
Measures of Disease Frequency (2)

• Incidence quantifies the of new events or
  cases of disease that develop in a population at
  risk during a specified time interval
• Cumulative incidence new cases or events
  during specified time period/ total pop at risk
  note assumes that everyone in study for
  the whole time period
• Incidence rate new cases or events during
  specified time period/ total person-time of
  observation note, accounts for different amounts
  of time per person in the study

9
Descriptive Studies (1)

• Describe patterns of disease occurrence in
  relation to variables such as person, place, time
• The data provided by descriptive studies are
  essential for public health administrators
• Knowledge of which populations or subgroups are
  most or least affected by disease allows the most
  efficient allocation of resources and the
  targeting of particular segments of the
  population for education/ prevention programs
• Descriptive studies can lead to hypothesis
  formation/ research questions

10
Descriptive Studies Person

• Address the question Who is getting the disease/
  event?
• The characteristics of person must include age
  and sex can also include religion, marital
  status, personality type, race, education level,
  income level, occupation, or others
• The interpretation of age distribution of disease
  is not always easy a change in the age
  distribution of disease over time can be due to
  many factors, such as improvements in diagnostic
  technology, or the discovery of new treatments
  for the disease

11
Descriptive Studies Place

• Address the question Where are the rates of the
  disease highest and lowest?
• Descriptive characteristics related to place can
  provide major insights into disease etiology
  (e.g., specific cancers in certain geographic
  areas)
• Geographic comparisons of disease frequency can
  be made between countries or between regions
  within a single country
• Descriptive data can be presented very
  efficiently in a pictorial manner, with maps,
  etc.

12
TB Case Rates, USA, 1999
D.C.
3.6 - 6.4
6.4 (national average)
Rate cases per 100,000
13
Descriptive Studies Time

• Address the question When does the disease
  (event) occur? Is the frequency of the disease
  at present different from the corresponding
  frequency in the past?
• Changes in disease rates over time can signal an
  epidemic or introduction of the causal agent
• Cyclic changes, such as seasonal patterns, are
  very valuable

14
Exercise Quantitative Analysis

• Use of Epi-INFO to analyze National TB Registry
  Data from S.African Country
• Data from two districts
• District 03
• District 15
• Interpret the data what is happening with TB in
  these areas? What can you, as a TB program
  manager, recommend?

15
Exercise What to Analyze?

• Descriptive statistics
• Person
• Who has TB gender/ age group--what does that
  imply about transmission in these districts? Is
  the TB pulmonary/extrapulmonary?
• Place
• Are there differences by district?
• Time
• Are there seasonal patterns to the disease?
• Has TB remained the same year after year?