Sampling in Epidemiology

Sampling in Epidemiology
EPIDEMIOLOGY
SimonPeters
Models of Epidemiology
• Host, agent, environment
• Person, place and time
• Exposure-outcome
• Cause-effect
• Observational vs. experimental
• Mathematical and statistical
Models in Epidemiology
• Common features across models
– Sampling
– Measurement
• Why?
– Unable to study all members of a population
– Reduce bias
– Save time and money
– Measurements may be better in sample than in
entire population
– Feasibility
• Definitions
– Sampling unit – the basic unit around which a
sampling procedure is planned
• Person
• Group – household, school, district, etc.
• Component – eye, physiological response
– Sampling frame – list of all of the sampling
units in a population
– Sample – collection of sampling units from the
eligible population
• Probability (random) sampling
– Sampling in which each sampling unit has a
known and nonzero probability of being
included in the sample
• Replacement
– With replacement – sampling unit returned to
population before next sampling event
– Without replacement – sampling unit not
returned to population before next sampling
event
• Random Sample • Non-random Sample
– Simple random sample – Convenience sample
– Stratified random – Systematic sample
sample – Consecutive sample
– Cluster sample – Quota sample
– Adaptive cluster – Volunteer sample
sample – Capture-recapture
– Multistage sample
• Which sampling design is best?
– Choose the method that gives the greatest

degree of accuracy and precision for a given
cost.
• Simple random sampling
– Each sampling unit has an equal chance of
being included in the is sample
– In epidemiology, sampling generally done
without replacement as this approach allows for
a wider coverage of sampling units, and as a
result smaller standard errors
– Advantages
• Simple process and easy to understand
• Easy calculation of means and variance
– Disadvantages
• Not most efficient method, that is, not the most
precise estimate for the cost
• Requires knowledge of the complete sampling frame
• Cannot always be certain that there is an equal
chance of selection
• Nonrespondents or refusals
– Estimate hemoglobin levels in patients with sickle cell
anemia
1. Determine sample size
2. Obtain a list of all patients with sickle cell anemia
in a hospital or clinic
3. Patient is the sampling unit
4. Use a table of random numbers to select units
from the sampling frame
5. Measure hemoglobin in all patients
6. Calculate mean and standard deviation of sample
• Systematic sampling
– The sampling units are spaced regularly
throughout the sampling frame, e.g., every 3rd
unit would be selected
– May be used as either probability sample or not
• Not a probability sample unless the starting point is
randomly selected
• Non-random sample if the starting point is
determined by some other mechanism than chance
• Systematic sample
– Advantages
• Sampling frame does not need to be defined in
advance
• Easier to implement in the field
• If there are unrecognized trends in the sample
frame, systematic sample ensure coverage of the
specturm of units
– Disadvantages
• Variance cannot be estimated unless assumptions
are made
• Systematic sampling
– Estimate HIV prevalence in children born
during a specified period at a hospital
1. Impossible to construct sampling frame in
advance
2. Select a random number between some pre-
specified bounds
3. Beginning with the random number chosen, take
every 5th birth and measure for HIV infection
• Stratified random sample
– The sampling frame comprises groups, or
strata, with certain characteristics
– A sample of units are selected from each group
or stratum
– Advantages
• Assures that certain subgroups are represented in a sample
• Allows investigator to estimate parameters in different strata
• More precise estimates of the parameters because strata are more
homogeneous, e.g., smaller variance within strata
• Strata of interest can be sampled most intensively, e.g., groups with
greatest variance
• Administrative advantages
– Disadvantages
• Loss of precision if small number of units is sampled from strata
– Assess dietary intake in adolescents
1. Define three age groups: 11-13, 14-16, 17-19
2. Stratify age groups by sex
3. Obtain list of children in this age range from
schools
4. Randomly select children from each of the 6
strata until sample size is obtained
5. Measure dietary intake
• Cluster sampling
– Clusters of sampling units are first selected
randomly
– Individual sampling units are then selected
from within each cluster
– Advantages
• The entire sampling frame need not be enumerated
in advance, just the clusters once identified
• More economical in terms of resources than simple
random sampling
– Disadvantages
• Loss of precision, i.e., wider variance, but can be
accounted for with larger number of clusters
– Estimate the prevalence of dental caries in
school children
1. Among the schools in the catchment area, list all
of the classrooms in each school
2. Take a simple random sample of classrooms, or
cluster of children
3. Examine all children in a cluster for dental caries
4. Estimate prevalence of caries within clusters
than combine in overall estimate, with variance
• Multistage sampling
• Similar to cluster sampling except that there
are two sampling events, instead of one
– Primary units are randomly selected
– Individual units within primary units randomly
selected for measurement
• Multistage sampling
– Estimate the prevalence of dental caries in school
children
1. Among the schools in the catchment area, list all of the
classrooms in each school
2. Take a simple random sample of classrooms, or cluster of
children
3. Enumerate the children in each classroom
4. Take a simple random sample of children within the
classroom
5. Examine all children in a cluster for dental caries
6. Estimate prevalence of caries within clusters than combine
in overall estimate, with variance
• Convenience sample
– A non-random collection of sampling units
from an undefined sampling frame
• Advantages
– Convenient and easy to perform
• Disadvantages
– Not statistical justification for sample
• Convenience sample
– Case series of patients with a particular
condition at a certain hospital
– “Normal” graduate students walking down the
hall are asked to donate blood for a study
– Children with febrile seizures reporting to an
emergency room
Investigator decides who is enrolled in a study
• Consecutive sample
– A case series of consecutive patients with a condition
of interest
– Consecutive series means ALL patients with the
condition within hospital or clinic, not just the patients
the investigators happen to know about
• Advantages
– Removes investigator from deciding who enters a study
– Requires protocol with definitions of condition of
interest
– Straightforward way to enroll subjects
• Disdavantage
– Non-random
• Consecutive sample
– Outcome of 1000 consecutive patients
presenting to the emergency room with chest
pain
– Natural history of all 125 patients with HIV-
associated TB during 5 year period
Explicit efforts must be made to identify

and recruit ALL persons with the condition
of interest
• Capture-recapture sampling
– A non-random method of sampling that relies
on lists of sampling units obtained from
multiple sources.
– The overlap in the lists allows one to estimate
the number of individuals not ‘captured’
• Uses of this method
– Estimate parameter when incomplete
information is available from  2 sources
– Refine of prevalence or incidence estimates
from population surveys
– Assess completeness of event reporting
– Derive plausible upper and lower limits on total
population affected
• Advantages
– Does not require random sample
– Can give more precise estimate of parameter than
probability sample
– Easy to perform in the field
– Useful in estimating events in difficult to access
populations
• Disadvantages
– Analysis of lists may be complicated
– Need to be able to match individuals across lists
– Assumptions regarding probability of being listed by a
source
– Unfamiliar to epidemiologists
• Capture-recapture
– Estimate the number of AIDS cases among IDUs in a
city
1. From hospital and clinic records obtain lists of persons
with diagnosis of HIV/AIDS during study period
2. Determine IDU status
3. Identify people who appear on multiple lists
4. Use nested log-linear models to estimate the number of
IDUs with AIDS not captured by the different lists
5. Use the list of reported cases and estimate of non-
reported cases to obtain overall estimate of the number
of IDUs with AIDS (with confidence intervals)

Sampling in Epidemiology

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sampling in Epidemiology

Uploaded by

Copyright:

Available Formats

Sampling in Epidemiology

– Choose the method that gives the greatest

Explicit efforts must be made to identify

You might also like