Epidemiology
blank.
L8_TypesofEpiStudiesADA1.pptxTypes of Epi Studies
1
Overview of Lesson
Experimental Studies
Observational Studies
Descriptive
Individual Level
Group Level
Analytical
Recall that the purpose of epi studies is to identify causes (risk factors) for diseases so as to elucidate disease etiology, a fancy way of saying determining what causes a disease. Epi studies are done to explore new ideas about how and why disease occurs, and to clarify a possible cause-and-effect relationship that seems unclear.
The hypothesis about these ideas and possible relationships are typically based on anecdotes, clinical observations, case studies, and animal studies. Because these types of observations have no "controls" (with the exception of animal studies), the conclusions that are made cannot be said to be "true" with any measure of confidence. The question here is how to test these hypothesis in a way that is logical, feasible, and ethical.
There are two general types of research designs: experimental and observational. The experimental approach is probably the more familiar method of testing hypothesis since it is the basic model of investigation for other sciences.
2
Experimental Studies
Experimental: basic model of investigation for most sciences
Types of experimental studies in epi
Clinical Trial
Field or Community Trial (quasi-experiemental)
3
With experimental (intervention) studies, the investigator manipulates the exposure at an individual level (or at a group level – such as with fluoride in a water system -- in quasi-experimental designs) and then observes the result of this E manipulation on D. For example, let's suppose an investigator wants to determine whether substance X -- a highly carcinogenic agent -- is associated with brain cancer. In an experimental study, an investigator can take a litter of genetically identical rats, and give one of the two randomly assigned groups of rats substance X, then record the frequency with which cancer develops in the two groups.
When done well, experimental studies give us a very good idea of the truth about whether E causes D; however, they create an "unnatural" setting which can sometimes make the results misleading. For example, while we can learn a great deal from animal studies, humans are a whole different animal… Further, it is ethically impossible to expose people to harmful agents, so some E D pairs cannot be tested. Imagine trying to conduct the aforementioned substance x study on humans….
Epi purists contend that true epi studies are all observational and because experimental studies investigate potential treatment or cures, they really do not meet the basic purpose of epi studies. However, because experimental studies can be conceptualized as interfering with the disease development process, they can disclose information on disease etiology. Thus, experimental studies are considered by some (the non-purists!) to be epidemiologic. A classic example is that of a ship's captain giving half of his crew limes and the other half no limes and then observing the effects on the development of scurvy.
While this example makes experimental designs appear simple, to have a well-designed experimental study is very difficult and requires a great deal of careful planning (and money!). The specifics of how to do conduct experimental studies is a course of study in and of itself (research methodology). For that reason, we will NOT be covering experimental designs in this course.
Please note that clinical trials and community trials are experimental designs, as is “evidence-based medicine.”
Observational Studies
Describes “how things are”
Shows association, not causation
Uses “naturally segregated” groups of people to compare/contrast
Exposed/Unexposed
Diseased/Health
4
“True” epi studies are observational in that they study people who are already segregated into groups on the basis of either an exposure or disease status. They study these groups -- the exposed and the unexposed, or the diseased and the healthy -- as we find (or observe) them; that is, as they occur naturally. The study is observational because the differences between these groups are observed and analyzed as naturally occurring events , not events created experimentally through a manipulated intervention where some group are given E and some are not.
The difficulty with observational studies is that the observed groups usually differ in some characteristics in addition to the E of interest. This can confound the analysis, making it harder to determine the “truth.” We will learn more about this in a different lesson.
For the purpose of this course, it is necessary that you are able to identify the basic differences between experimental and observational studies.
Observational Designs
Descriptive
Characterizes a population
Generates hypothesis
Analytical
Tests hypothesis
Observational designs can be either descriptive or analytical. Up until this point in the semester, we have been studying descriptive epidemiology, where we describe disease occurrence within populations and sometimes compared different populations with one another. Descriptive studies are great at generating hypothesis.
Analytical epi looks more critically on the role of E on D, and tests hypothesis.
5
Descriptive Study Designs
Conducted at Individual Level
Case reports
Case series
Cross-sectional studies
Conducted at Population Level
Ecological studies
Time series
6
There are several different descriptive study designs, some of which are conducted at the individual level and others which are conducted on groups of people.
Individual Level: Case Report
Cases of unusual diseases or symptoms are described in terms of personal characteristics
Useful in
The recognition of new disease
Alerting the medial or general community of a possible problem
7
A case report is a careful and detailed report of a single patient/individual who has a disease and exposure of interest. It details the progress of symptoms and signs, response to treatment, etc.
Case reports are most commonly used when a physician sees a person with a rare disease and/or an uncommon exposure. It is most useful for disease recognition and raising awareness among other practitioners. They are cheap and easy to do. As they have no control or comparison group, though, the validity of the E D relationship is weak; that is, we cannot say with any confidence that E and D are related.
Individual Level: Case Series
An expansion of the case report
Several or many people with the same disease(usually odd) and the same exposure
Usually describes the characteristics of a number of patients
Distribution of cancer by age group,
sex and occupation, etc.
8
A case series is, in essence, an expanded version of a case report. They tell of multiple incidents with unusual patient disease and/or exposure status. Case reports and case series are useful in generating ideas about diseases and/or exposures and for raising awareness in public health and medicine that a problem exists. Classic case reports and series which developed important public health matters include Legionnaire's disease, Hanta Virus, and thalidomide
The biggest concern with case reports and case series is that there is no comparison population (group) which can lead to incorrect assumptions regarding the E D relationship.
Individual Level: Cross Sectional Study (aka Prevalence Study)
Determines disease and exposure status as they exist in a defined population at one specific point in time
Snapshot of
Current disease status of a population
Current exposure
9
Cross-sectional Studies: by means of a survey, current Es and current Ds are ascertained concurrently at a single point in time. The problem here is a lack of knowledge of the temporal relationship between E and D. That is, there is no way of knowing which came first: E or D. This makes this design weak for studying disease etiology.
Often, the exposure of interest is a personal characteristic (e.g., skin color, blood type, etc.) that is not modifiable. If it is a "lifestyle" exposure, however, that is being measured (e.g. diet, exercise levels, etc.), the current exposure may -- or may not -- be a reasonable surrogate for historical exposure (which is typically more relevant to D today).
Cross sectional studies are useful for studying disease or conditions that are not rapidly fatal, have no clear point of onset, and are transitory -- hypertension, osteoporosis, arthritis, diabetes, asthma, bronchitis and allergies for example. They are not appropriate, however, for rare diseases or diseases of short duration, because they are a snapshot of what is happening at a single moment in time.
Group Level: Ecological Studies
Unit of observation is the population or community.
Population can be defined many ways
Disease rates and exposures are measured in each of a series of populations and their relation is examined.
E and D info often abstracted from published statistics and
does not require expensive or time consuming data collection.
10
An ecological study is an observational study in which at least one variable is measured at the group level. Group data on E may be environmental (e.g., average rainfall), lifestyle related (e.g., tobacco sales data), diet-related (e.g., Dept. of Agriculture data on foods sold or consumed, etc.). Group data on D is typically from morbidity and mortality reports. Examples of group-level measures include the rate of cancer incidence, the mean level of hypertension, the average sunlight exposure at specific geographic location, etc.
In ecological studies the unit of comparison is the group or population (e.g., countries, age groups, racial groups, etc.). The variation in measures of effect between groups may be “true” differences, or they may be explained by other factors that haven’t been measured. Thus, it is very problematic when trying to infer the E D relationship from ecological studies. Thus, ecological studies are better at generating hypotheses than in answering them…
Ecological Fallacy
Situation in which a relationship between an exposure and a disease at the ecological (group) level, does not hold true at the individual level
Generally, ecological associations are stronger than individual ones
11
Ecological studies are prone to ecological fallacy, which happens when an inference is made about an individual based on aggregate data for a group.
Here’s a great example to illustrate taken from: http://jratcliffe.net/research/ecolfallacy.htm
Researches looked at the aggregate data on income for a neighborhood of a city, and discover that the average household income for the residents of that area is $30,000. To state that the average income for residents of that area is $30,000 is true and accurate. No problem there. The ecological fallacy can occur when the researcher then states, based on this data, that people living in the area earn about $30,000. This may not be true at all, and may be an ecological fallacy.
A closer examination of the neighborhood might discover that the neighborhood is actually composed of two housing estates: one of a lower socio-economic group of residents, and one of a higher socio-economic group. The poorer part of town residents earn on average $10,000 while the more affluent citizens can average $50,000.
When the researcher stated that individuals who live in the area earn $30,000 (the mean rate), this did not account for the fact that the average in this example is constructed of two disparate groups, and it is likely that not one person earns $30,000.
Assumptions made about individuals based on aggregate data are vulnerable to the ecological fallacy. This does not mean that identifying associations between aggregate figures is necessarily defective, and it doesn't necessarily mean that any inferences drawn about associations between the characteristics of an aggregate population and the characteristics of sub-units within the population are absolutely wrong either. What it does say is that the process of aggregating or disaggregating data may conceal the variations that are not visible at the larger aggregate level, and researchers, analysts and crime mappers should be careful.
Time Series (aka Surveillance)
Looking at a single population, but the unit of observation is now defined by time, not geography
Looking at changes in community exposure and community disease
No individual measures
12
Many diseases show remarkable fluctuations in incidence over time. Rates of acute infection can vary appreciably over a few days, but epidemics of chronic disorders, such as lung cancer and coronary heart disease, evolve over decades. If time, or secular trends, in disease incidence correlate with changes in a community's environment or way of life, then these trends may provide important clues to disease etiology. Ex: Portion sizes have increased and so has obesity.
Time series – or epidemiologic surveillance -- is the ongoing systematic collection, recording, analysis, interpretation, and dissemination of data reflecting the current health status of a community or population.
Analytical Study Designs
Follow-up (cohort)
Case-Control
13
You will have several lessons that focus on these analytical study designs, and how to conduct them, so I’m going to leave it at that for right now. This lesson is DONE! =)
