Design Strategies and Statistical Methods in Descriptive Epidemiology Flashcards
Study design
program that directs the researcher along the path of systematically collecting, analyzing, and interpeting data
Descriptive epidemiology
organizes, summarizes, and describes epidemiologic data by person, place, and time; characterizes the distribution of health-reltaed states/events by who, what, when, where
Types of descriptive studies
ecologic studies
case reports
case series
cross-sectional surveys
Ecologic
involves aggregate data and makes comparisons across large populations
Case reports
involves profile of a single individual
Case series
involves small gorup of patients with similar dx
Cross sectional survey
conduted over a short period of time with no follow-up period and the unit of analysis is the individual
Serial survey
cross-sectional surveys that are routinely conducted
Common serial surveys include
US census, Behavioral Risk Factor Surveillance System
Types of data
Nominal
Ordinal
Discrete
Continuous
Nominal examples
sex, diseases (yes/no), race, marital status, education status
Ordinal examples
preference rating (agree, neutral, disagree)
Discrete examples
number of cases
Continuous examples
dose of ionzing radiation
Nominal
Categorical – unordered categories
Ordinal
Categorgical – ordering informative
Discrete
Quantitative – integers
Continuous
Quantitative – values on a continuum
Ratio formula
X/Y x 10^
Ratio
the values of x and y are independent; x is not contained in y; rate base is 1
Proportion
x is contained in y; typically expressed as a %; rate base is 100
Rate
a proportion with the addition that it represents the number of health-related states or events in a population over a specified time period
Rate equations
incidence rate mortality rate attack rate person-time rate secondary attack rate prevalence proportion case-fatality rate
Incidence rate
new cases / population at risk x 10n
Mortality rate
deaths during a period/population from which death occured x 10n
Attack rate
new cases occuring during a short-time period / population at risk x 10n
Person-time rate
number of cases during observation period / time each person observed totalled for all persons x 10n
Secondary attack rate
new cases among contacts of primary cases / poplation at beginning of time-period - primary cases x 10n
Prevelenace population
new and existing cases of disease / total study of population x 10n
Attack rate (cumulative incidence rate)
diseases or events that affect a larger group of the population
Crude rate is calculated without any _____ and is limited upon comparison between_____ because of _____.
restrictions (age or sex) on who is counted in the numerator or denominator; subgroups of the population; confounding factors (age-distribution between groups)
Age-adjusted Rate
adjusting for differences in the age distribution among a population to eliminate confounding factors
Two methods for calculating age-adjusted rates
direct and indirect
SMR
standarized morbidity ratio
SMR formula
SMR = Observed/expected
confounding
the distortion of the association between an exposure and health outcome by an extraneous, third variable
confounding examples
age
sex
educational level
smoking
confounder (confounding variable/lurking variable)
an extrinsic variable that distorts the outcome of what is being studied
spurious association
caused by a third factor (confounder) that is not apparent at the time of examination
