Descriptive Studies
(3 criteria)
Always observational
Researchers do not manipulate exposure
Mainly concerned with person, place, and time
Descriptive Studies (limitations)
can not infer causality
Seeks to describe distribution of variables (cannot be used to test for the presence of valid statistical association
Descriptive Studies are used to (5 things)
Generate hypotheses
Monitor public health policy
Conduct surveys
Planning, assessment
Guide further studies.
Descriptive studies measure
Social, economic and health conditions of population such as birth and death information (vital statistics).
Difference between Descriptive and Analytical Studies
Descriptive can provide a hypothesis
Analytical can establish more of a causal link
Descriptive Studies summary
Three types of Descriptive Studies
Ecologic Study (also known as Correlational Study).
Cross-Sectional Study(also known as Prevalence Study).
Case Study / Case Series Study.
Ecological Study
Examine populations / groups rather than individuals.
Focus on the associations between exposures and outcomes in the selected populations.
Data is collected in the past.
Advantage of being quick to perform and generating new hypothesis for exploration of risk factors.
Disadvantage is that results for community/population may not infer to the individual.
Eco studies are
good to get general picture and infer additional studies for risk factors but care must be taken to not infer population results to the individual
Ecological fallay
Inaccurate inference from population or community results to the individual
Case Study definition
Report by individual or groups of individuals with specific conditions; no control group used.
Qualitative description of a person, group or event. Researcher studies individual cases and seeks to explain causality.
Purpose of a case study
Often used to identify the beginning or presence of an epidemic or unusual outbreak.
To formulate hypothesis for further study; suggest causality.
Limitation of a case Study
Cannot be used to establish formal statistical association.
Since case report is often examination of single cases and therefore limits strong conclusion, case series are often used.
Prospective Case Study
Subjects included in study as they occur or are found meeting the criteria for inclusion.
Retrospective
Subjects are selected because historical records meet established criteria.
Case (series) study
Can also be prospective or retrospective by tracking groups of patients with a known exposure and/or treatment into the future to examine outcome of examine records for exposure and outcome
Consecutive
Results include all members of the study
non consecutive
Only select participant included in the results
Problem with case study/series
Case study and case series studies are subject to confounding by selection bias which limits inference on causality
Cross Sectional Study
Observes a population of interest at a single point in time or during a specific time interval.
Can be used to survey or assess the health status of a population by gathering information from individuals such as history, habits, knowledge, and behavior.
Advantage of being able to simultaneously assess exposure and disease status (outcome) in a population.
Relatively inexpensive and fairly good generalizability.
Limitation is that it is difficult to determine when exposure occurred
Purpose of a cross sectional study
Examines the relationship between diseases (or other events) and possible associated factors as they exist in the population at one point in time.
Measures exposure prevalence to disease prevalence.
Limitations of a cross sectional study
Healthy Worker Effect (workers seem to have lower morbidity/mortality rates because left work due to disability or death and therefore not included in study).
Preponderance of prevalent cases of long duration (subjects who have had condition for a long time and have “acclimated” coal miners).
Most common form of a cross sectional study
Survey
Examples
BRFSS - Behavioral Risk Factor Surveillance System
NHANES - National Health and Nutrition Survey
Most opinion and political polls.
Sampling schemes for cross sectional study (2 types)
PROBABILITY SAMPLE & NON-PROBABILITY SAMPLE
Definition of a probability sample and the 3 types
Is sample where everyone in the population has a “non-zero” chance of being included (in other words, everyone has a chance of being included!)
Simple Random Sample:
Systemic Sample:
Stratified Sample:
simple random sample
Everyone in population has equal chance of being selected (community survey, census, etc.).
systematic sample
Subjects are selected based on some established scheme
(every 5th person, every second house, LQAS, one school’s 9th graders, etc.).
stratified sample
A pre-established (calculated) number of subjects are randomly selected from subsets of the total population (CNA).
Non probability sample definition and 3 types
Probability sampling is best for large-scale studies, non-probability approaches are best for complex or rare events.
Judgement Sampling
Convenience or accidental sampling
Quota sampling
Judgment Sampling
Researchers choose subjects based on who they think would be most appropriate for the study; used when specific issue is being studied.
Convenience or acidental sampling
Samples are chosen based on accessibility (friends, mall shoppers, etc.); prone to bias.
Quota sampling
A quota is established and subjects chosen until quota is met (65% male); often not representative.
1. What is an example of ecologic fallacy?
A. Inaccurate inference from population or community results to the individual.
B. Inaccurate inference from the study population to the general population.
C. Inaccurate inference from one study population to another study population
D Both B and C
E None of the above
A
What is/are the limitation(s) of a cross-sectional study?
A. It only measures incidence
B. The Healthy-Worker effect
C. Preponderance of prevalent cases of long duration
D. Both B and C
D
3. What is the main difference between Descriptive and Analytic studies?
A Descriptive studies only examine individuals
B Descriptive studies are experimental
C Descriptive studies do not manipulate exposure
D None of the above
C
4. Which of the following is not a descriptive study?
A Cross-sectional
B Case series
C Ecological
D Experiment
D
5. A study examining the effects of a nearby waste disposal site sampled every fifth house in a series of neighborhoods. This is an example of what type of sampling?
A Systematic sampling
B Stratified sampling
C Simple random sampling
D Judgment sampling
B
6. Non-probability samples are best used for:
A Rare events
B Complex studies
C Both A and B
D None of the above
C
7. What are case-studies/case series MOST used for?
A To detect an outbreak
B To determine causality
C To examine the burden of disease in a population
D To establish formal statistical association
A
8. What study design can incorporate consecutive and non consecutive participants?
A Case series
B Ecological
C Cross-sectional
D None of the above
A
9. What is/are the advantage(s) of a cross-sectional study?
A Good generalizability
B Inexpensive
C Can infer causality
D Can infer temporality
E Both A and B
E
10. Which of the following is a cross sectional study?
A A study describing the prevalence of smoking among high school students in Harris county.
B To determine the association between smoking and lung cancer, a researcher compared the smoking histories of 500 subjects with lung cancer and 1000 subjects without lung cancer.
C Two groups of elderly subjects – one group vaccinated, the other group unvaccinated were studied to determine the long-term effectiveness of influenza vaccines in elderly people.
D None of the above.
A
11. To study the association between knowledge and behaviors related to HIV in high school students, a questionnaire to assess knowledge about prevalence, causes and individual susceptibility to HIV and practice of risky sexual behaviors was administered.
This is an example of which type of study?
A Case series
B Case-control
C Cross-sectional
D None of the above
C
A study that examines the death rates from colon cancer in each of the 50 U.S states in relation to the average percentage of residents in each state undergoing colonoscopy screening. This is an example of :
A Ecologic
B Cross sectional
C Case control
D Cohort
A
13. A study comparing the prevalence of cardiovascular disease in the US with the prevalence of cardiovascular disease in Japan is an example of _____
A Case series
B Case-control
C Cross-sectional
D Ecological Study
D
14. Which of the following is a cross sectional study?
A A study describing the prevalence of depression among adult patients with Rheumatoid Arthritis
B A study describing prevalence of developmental disorders among children with Autism.
C None of the above
D Both A and B
D
15. A study describing the 5cases of West Nile Virus disease in Tarrant County in 2012 is an example of _____
A Case series
B Case-control
C Cross-sectional
D Ecological Study
D
15. A researcher set out to investigate the relationship between three major allergic diseases, asthma, allergic rhinitis (AR), and atopic dermatitis (AD), and psychological and behavioral problems in preschoolers. He conducted a survey using a modified International Study of Asthma and Allergies in Childhood questionnaire to determine the prevalence of symptoms and diagnosed allergic diseases, and a Korean version of the Child Behavior Checklist to assess internalizing, externalizing, and sleep problems among 780 preschoolers. This is a ____study design.
A Case series
B Case-control
C Cross-sectional
D Ecological Study
C
16. A study describing the prevalence of childhood obesity in Fort Worth is a Cross-Sectional study.
True
False
True