Lecture #12 - Cross-section Studies and Ecological Studies Flashcards
True or False:
Cross-sectional studies are described as Interventional and Descriptive
False,
Observational and Descriptive
Which has a higher strength of evidence?
Cross-Sectional or Cohort?
Cohort
Cross sectional studies tend to look at information gathered at one point in time or time frame. This is commonly called what?
Snap-shot in time
For Cross Sectional Studies, state how they looks at:
Person:
Place:
Time:
Person: All of interest (exposure and disease)
Place: Defines geographical areas
Time: Specific time of study (specific)
What is the major advantage of cross sectional studies?
They are quick and easy
Easier/Quicker IRB approval
Which of the following avenues are most commonly and appropriately utilized in Cross-sectional studies?
A. Scanning-electron microscope
B. Population Surveys
C. Direct patient interview
D. Patient Follow up data
B. Population Surveys
What are the disadvantages of cross sectional studies?
1) May be difficult to sort out factors associated with risk
of disease from factors associated with survival (such
as treatment & severity)
2) Difficult to study diseases of low frequency
(Prevalence of disease in population is proportional to
the incidence of the disease times its duration)
3) Problems in determining temporal relationship of
a presumed cause & effect b/c exposure & disease histories are
taken at the same time
What is Neyman Bias?
selective survival
States how the study would not be effective for acute or rapidly fatal diseases (people with acute diseases die quickly and are only available for a short period of time)
What are the two ways to approach a cross-sectional study?
1) Collect data on each member of the population
2) Take a sample of the population & draw inferences
to the remainder (MORE COMMON)
What is the difference between a probability sample and a non-probability sample?
Probability samples
o Every element in the population has a non-zero
probability of being included in sample
Non-probability samples:
Based on a sampling plan that does NOT has a non-zero probability
** not appropriate for Cross-sectional studies because reliability of results cannot be evaluated.
Define:
Simple Random samples
Systematic samples
Stratified samples
Simple Random: List of population, randomly selected
Systematic samples: List of population, define variable by which you will select (ie all persons with last name A-L or sampling every 20th person) - if there is some order to the list picking every 20th person could introduce selection bias
Stratified Samples: Mutually exclusive & exhaustive strata – age or
socioeconomic groups, divide population into relevant subgroups. take simple random sample from each subgroup
Cross-sectional studies use what two approaches to collection of NEW information?
o Questionnaires/Surveys
Either directly from patients or their caregivers/treating
institutions
o Physical assessments (which might involve laboratory,
clinical, or psychological tests)
what is unique about NHANES?
National Health and Nutrition Examination Survey
RV’s go to area - draw blood, EKGs, throughout the US
THINK EXAMINATION AND SURVEY
Assesses the health & nutritional status of adults & children
o Combines interviews & physical examinations
What is unique about NHIS?
National Health Interview Survey
Principal source of information on health of the civilian, non-institutionalized population (NOT PEOPLE IN HOSPITALS)
Data are collected through a personal household interview
Consists of a set of core questions that remain largely unchanged & a set of supplements used to respond to public health data needs as they arise
What is unique about NAMCS?
National Ambulatory Medical Care Survey
Focuses on ambulatory medical care services in the United States (direct patient care visit surveys)
SEE DOCTOR, go home, get records at hospital
What is unique about NHAMCS?
National Hospital Ambulatory Medical Care Survey
A national survey designed to collect data on the utilization and provision of ambulatory care services in hospital emergency (non-inpatient) and outpatient departments
H is abbrev. indicates Hospital emergency
What is unique about BRFSS?
Behavioral Risk Factor Surveillance System
Collects information on health risk behaviors, preventive health practices, and health care access primarily related to chronic disease and injury
**Interviewed by telephone, Largest landline telephone health survey in the world **
Youth BRFSS conducted by questionnaire in schools
True or False:
Observational studies are described as Interventional and Descriptive
False, Observational and descriptive
What is the major difference between Cross sectional studies and Ecological Studies?
Ecological studies look at populations as a whole, cross sectional look at individuals.
Ecological: Informs researchers about the forces which act on
whole populations, which may be in conflict with those
that act on individuals
The following study would be an example of what kind of study:
Pregnancy smoking rates among non-Hispanic black women and white women in Kansas City are associated with same-race zip-code population density
Rates for black women decrease with increasing zip
codes integration, while rates for whites increase with
increasing zip code integration
Ecological
Which of the following do Ecological studies use?
incidence rates, prevalence, mortality rates
all. Can do all at the same time on a large population.
Level of exposure for each individual is UNKNOWN
True or false: Ecological studies provide results that show causation between two sets of data
False. Do not show causation, only correlations.
Clinical Case-Series is… (design)
o Data derive from practice of individual physicians or health care organizations
o Effectively a Case-Registry
Population Case-Series is…
o A clinical Case-Series in which data is complete for a defined geographical area for which population is known
extra cases not seen by physician ARE included (coroner)
provides National or even International perspective on disease
NO measure on the individual
The following are an example of what?
1% reduction in unemployment raises CHD
mortality 0.75% or almost 3,900 additional deaths
1% decrease in national unemployment is
associated with 0.4% rise in mortality
Smoking & excess weight decline during temporary
economic downturns while leisure-time activity
rises
Ecological Outputs
Ecological Fallacy states….
the association found with aggregate data may not
likely apply to individuals
Atomistic Fallacy states…
wrongly assuming, from observations on causes of
disease in individuals, that the same forces apply to
whole populations (Opposite of ecological fallacy)
The following is an example of which fallacy?
At the individual level a high income (a marker of
material success) is often assumed to be associated
with a lower rate of suicide (or better mental health)
The opposite seems to be true in KC…
Atomistic Fallacy
Is Ecological Fallacy the same as External Validity?
No. They are the same in the sense that in essence you are looking to see if everybody else is the same as what is found BUT the difference is:
1) Ecological Fallacy - always failing
2) External validity is specific to the study, so, does this treatment for cancer apply to others with that specific type of cancer. Ecological Fallacy tends to pop up when there is a very broad finding and trying to apply it to everybody
Homogeneity Fallacy states…
misrepresentation of population data from heterogeneous (diverse in character) populations
The following could be an example of which fallacy?
European-origin Caucasian populations in England
have a lower all-cause SMR than persons born in
the Indian subcontinent (often called South Asians)
Homogeneity Fallacy
While this is true, the highest SMR is actually
within the Irish-born individuals living in England,
who are often included in the larger Caucasian
population
Cross sectional is to individual as Ecological is to..
population
