|
Table of Contents
Basic
Epidemiology
Epidemiology
Patterns
of Health Events
Measuring Frequency
Types of Rates
Making
Comparisons over Persons, Places and Time
Testing
for Significant Change Over Time
Working
Around the Small Numbers Problem
Other Measures in
VistaPHw
Years of Potential
Life Lost
Life Expectancy
Epidemiology Links:
http://www.apha.org/public_health/epidemiology.htm:
Really good site from the American Public Health Association
Epidemiology
Study of the distribution and determinants of diseases and injuries in
human populations.
Analytical Epidemiology
Search for determinants of disease or injury
Hypothesis testing of cause-effect relationships through epidemiological research:
-case control studies
-cohort studies
-intervention studies
Descriptive Epidemiology
Search for the pattern and frequency of health events in a population
Why do descriptive epidemiology?
-because the PHIP says we should
-monitor population health status
-define public health priorities
-target screening and intervention efforts
-generate hypothesis for analytic studies
Patterns of
health events relative to Person, Place, and Time
Patterns Relative to Person
Who has the greatest risk of getting a disease or other health outcome?
-Age
-Education
-Gender
-Race
-Occupation
-Income
-Behavior
-Ethnicity
Patterns Relative to Place
Where is the health outcome occurring?
-Statewide
-County level
-Subcounty level
Patterns Relative to Time
Is the frequency of the health outcome changing over time?
-Annual Trends
-Seasonal Occurrence
-Daily or hourly changes
Measuring Frequency
Counts and Rates
Counts
Number of events in the source population
Anatomy of a Rate
Number of events x Unit of Population (e.g. 100,1000,100,000)
Population "at risk" of experiencing the event
Rates
Events adjusted for the size of the source population in which they occurred (we expect to
find more cases in larger populations).
Facilitates comparison of disease frequency across different groups of people,
places, and time periods.
Types of Rates
Category-specific rates: age, gender, race
Crude rates
Age-adjusted rates
Category-Specific Rates
-Rates for categories of the population defined on the basis of particular
characteristics.
-Numerator AND denominator restricted to the group of interest.
Crude Rates
-Total number of events divided by TOTAL population - to be distinguished from
category-specific rates
-A rate that IS NOT "adjusted" or "Standardized" - in this sense,
category-specific rates can still be "crude."
-For mortality, crude rates can either be deaths from all causes OR cause-specific
(e.g., deaths from coronary heart disease)
The Problem with Crude Rates
Crude rates do not account for the underlying demographic differences between communities
(or between time periods) that can affect rates.
Two Solutions to the Problem
-Compare only age-specific rates - cumbersome!
-Construct age-adjusted rates: summary measures that account for differences in the
underlying age distributions of populations.
Age-Adjusted Rates
-Standardization
-Apply the observed age-specific rates to a single standard population.
1940 U.S. population
1970 U.S. population
2000 U.S. population
-The age-adjusted rate represents the hypothetical rate that would have been observed if
the population of interest had the same age distribution as the standardized population
-The age-adjusted standard is currently based on the year 2000 population.
Making
Comparisons Over Persons, Places, and Time
How stable are the rates?
When is a difference or change "statistically significant?"
Unstable Rates
-Rates based on small numbers of events are affected by random variation.
-The addition or deletion of a few cases dramatically affects the size of the rate.
-Comparisons across years or between areas are difficult to interpret when rates are
unstable.
Confidence Intervals
-The CI for a rate indicates what an expected range of random variability would be, based
on statistical assumptions.
-CIs are an indication of the stability of a rate - the wider the interval, the more
unstable the rate.
-95% CI means that there is a 95% probability that the CI contains the true, underlying
rate.
-80%, 90%, and 99% CIs are also options in VistaPHw.
CIs as a Test of Significance
Observed vs. Standared Rate:
-significant difference if the CI for the observed rate does not contain the standard (or
target) rate.
Two observed rates:
-significant difference if the CIs for the two rates do not overlap.
See also: http://www.doh.wa.gov/Data/Guidelines/ConfIntguide.htm
Testing for
Significant Change Over Time
Chi Square Test for Trend
Chi Square Test for Trend in
Proportions
-Tests whether an increase or decrease in a series of rates is random variation or a true
change occurring in the population.
-Not appropriate for age-adjusted rates or rolling averages
P-values
-For the chi square test, the p-value is the probability that the observed increase or
decrease is merely the result of chance.
-If the p-value is less than 0.05, there is less than a 5% chance that the observed
increase or decrease is due to random variation - trend is significant.
Testing for Significance
-Objective approach to evaluating difference or change
-Be careful about applying statistical tests without having really looked at your data
-Practical significance of a difference may differ from the statistical
significance.
Working
Around the Small Numbers Problem
Aggregation
Aggregating to Stabilize Rates
Grouping is used to increase the size of the numerator and denominator -
"stabilize" the rates.
-Combine across multiple years of data
-Combine across geographic areas (e.g., counties)
-Combine across groups of people (e.g., aggregate 5-9 and 10-14 age groups into a
5-14 age group)
Aggregation
-Numerator: sum of all events occurring during the period (events in year 1 + events in
year 2 + events in year 3)
-Denominator: sum of total population at risk during the period (population in year 1 +
population in year 2 + population in year 3)
-A significant limitation to combining across years, areas, groups is the loss of
information.
-Often a tradeoff between stability of rates and specificity of the
information. Consider the purpose of the analysis
Rolling Averages
-Used for graphical presentation to "smooth" trend lines.
-Overlapping aggregate rates.
-Not appropriate for statistical tests of significance.
Other Measures in VistaPHw
Years of Potential Life Lost (YPLL)
Life Expectancy
Years of Potential Life Lost (YPLL)
-Measures the impact of premature death (death that occurs before age 65)
-Counts the numbers of years between the age at death and age 65
Life Expectancy
-Average number of years a person can expect to live from a given age (e.g, birth).
-Based on the age-specific death rates in the year of interest.
-As mortality trends change over time, so does life expectancy.
-Longer life expectancy at birth reflects lower death rates especially among young people.
Table of Contents
|
