1 Page summary
HCS 420:
RESEARCH METHODS IN HEALTHCARE
Week 4
Chapter 7
Study Design I
Cross-Sectional and Cohort Studies
Chapter 7
Observational studies
https://www.youtube.com/watch?v=Z_OJzgkKe2A
What are Observational Studies?
Observational Study Designs
- Observational studies
“natural experiments” where the investigator does not manipulate treatment or exposure to disease, but rather observes natural variation
Advantages:
- Some risk factor / disease associations are not appropriate for experimental study (ethically)
- Can have high external validity (real-world)
Disadvantages
- Prone to bias
What are the purposes of observational studies?
Purposes of Observational Studies
- Descriptive: Examining the distributions of predictors and outcomes in a population.
- Analytic: Characterizing associations between predictor and outcome variables.
What are the types of observational studies?
Types of Observational Studies
- Main types are:
- Cross-Sectional study
- Cohort study
- Case-control study
What are Cross-Sectional Studies?
Categorized according to the time frame for making the study measurements.
In a cross-sectional study, all the measurements are made at about the same time, with no follow-up period
Steps
- Define the selection criteria
- Recruit a sample from the population
- Measure the current values of predictor and outcome variables (can be supplemented with historical information)
Examples
The National Health and Nutrition Examination Survey (NHANES): A program of studies designed to assess the health and nutritional status of adults and children in the United States.
It’s helpful for determining the prevalence of smoking in various demographic groups.
What are some strengths of observational studies?
Strengths
- Fast and inexpensive (no waiting for the outcome to occur)
- Avoids problems associated with loss to follow-up
- Can be used as the first step in other types of stud designs (e.g. cohort study or clinical trials)
What are some weaknesses of observational studies?
Weaknesses
- Difficult to establish causal relationships
- Difficult for investigating rare diseases
- Measures only prevalence (and not incidence); difficult to draw inferences about the causes, prognosis, or natural history of a disease
Cross-sectional studies
https://www.youtube.com/watch?v=t_Cu2fWn1fg
Cohort Studies
What are cohort studies?
Depends on the type..
Like cross-sectional study, cohort studies are categorized according to the time frame for making the study measurement
Types of cohort studies
(1) Prospective cohort studies
(2) Retrospective cohort studies
(3) Multiple cohort studies and External Controls
Prospective Cohort
The investigator begins by assembling a sample of subjects, measures characteristics in each subject that might predict the subsequent outcomes, and follows these subjects with periodic measurements of the outcomes of interest
Strengths of Prospective Cohort Studies
- Unlike cross-sectional designs, allows the calculation of incidence (i.e. the number of new cases of a condition occurring over time)
- Helps with the process of inferring the causal basis of an association since the levels of predictors are measured before the outcome occurs
- Predictor measurement is not influenced by the outcome or knowledge of its occurrence
Weaknesses of Prospective Cohort Studies
- Like other observational designs, causal inference is challenging and interpretation is hampered by the influences of “confounding variables”
- a variable that influences both the dependent variable and independent variable causing a spurious association
- Expensive and inefficient for studying rare outcomes
Retrospective Cohort
The investigator compares groups of individuals who are alike in many ways but differ by a certain characteristic (e.g. seniors who have dementia vs those who don’t) in terms of a particular outcome such as falling down)
Typically, the subjects are already assembled, baseline measurements have already been made, and the follow-up period has already taken place
What are the strengths?
Strengths of Retrospective Cohort Studies
- Most of the advantages of prospective cohort studies
- Less costly and time-consuming
- Everything is pretty much in place for the investigator to conduct his study
What are the weaknesses?
Weaknesses of Retrospective Cohort Studies
- Investigator has limited control over sampling and follow-up
- No control over the nature and the quality of the baseline measurements
- Existing data may be incomplete, inaccurate, or measured in ways that are not ideal for answering the research question
Multiple Cohort
Investigator begins with two or more separate samples of subjects (one group with exposure to a risk factor, and the other group with no exposure), define suitable cohorts with different levels of exposure to the predictor of interest, measures other predictor variables, follows up the cohorts, and assesses outcomes
Steps
- Select two or more cohorts from populations with different levels of the exposure (main predictor)
- Measure other predictors
- Measure outcome variables during follow-up
Strengths of Multiple Cohort Designs?
Strengths of Multiple Cohort Designs
- Similar strengths as other cohort designs
- May be the only feasible approach for studying rare exposures to potential occupational and environmental hazards
- Can be population-based and economical when data from a census or registry are used as the external control group
Weaknesses of Multiple Cohort Designs?
Weaknesses of the Multiple Cohort Designs
- Increased problem with confounding
- Differences between cohorts may not be measurable, creating problems in the interpretation of observed associations
Statistical Approaches to Cohort Studies
- Risk: N who develop the outcome / N at risk
- Odds: N who develop the outcome / N who do not develop the outcome
- Rate: N who develop the outcome / person-time at risk
- Incidence rates: The number of outcomes / person-time at risk