Epidemiology: Basics of Study Defining Epidemiology Design, Measures, and “…the study of the distribution and Limitations determinants of health-related states or events in specified populations and the Neal Simonsen application of this study to control of health problems.” LSU Health Sciences Center --Last, 1988, The Dictionary of Epidemiology
School of Public Health & Stanley S. Scott Cancer Center
Another Definition of Epidemiology Epidemiology Defining Itself
¾ “The science of making the obvious Etymological derivation: From the Greek obscure. ” “Epi” on/upon + --Anonymous Epidemiologist “demos” the people + “logos” theory or study of
Characteristics of Epidemiology
Concerned with the frequencies and types of illnesses and injuries in groups of people and the factors that influence their distribution.
1 Characteristics (Continued) Prevalence vs. Incidence rates Prevalence ¾ This implies that disease is NOT randomly distributed throughout a population, but ¾ Proportion of persons in a population who rather that subgroups differ in the frequency have a particular status (presence of disease or of different diseases. some other health-related condition) at either ¾ Knowledge of this uneven distribution can 1) A specified point in time be used to investigate causal factors and 2) A specified period in time thus to lay the groundwork for programs of prevention and control. ¾ Point vs. period prevalence ¾ Can similarly be used to study consequences of different treatments
Incidence Prevalence vs. Incidence rates
Incidence Prevalence (continued) ¾ The rate of new occurrences of a condition in a ¾ Amount of disease prevailing in a population population at a given time or within a given period = (New occurrences of a condition [=event] during a specified time period / the population during that same specified time period) What if we are interested in how quickly new cases are developing in a population? What if the condition of interest lasts a long time (years)? E.g., osteoarthritis
Exposure Study Designs In OR Genetic Background Epidemiology… OR Combination of Both Association Basic Considerations; ? Causation ? Fundamental Designs Disease or Other Outcome
How do we know if an observed association reflects a causal relationship?
2 Exploring Disease Etiology Dose/Treatment Experimental Study Design
Disease Exposed Cancer Exposure nonoccurrence assigned Not exposed OK Association
Unethical to perform experiments on Environmental ? Cancer people if exposure is harmful Exposure Not OK
http://www.epiet.org/course/Presentations/Presentations%202003/13-%20Cohort&Case%20control/13-%20Cohort&Case%20control_files/frame.htm
The next step in determining causation: Clinical observations (case series) Conducting Studies in Human Populations Ecological or Cross-Sectional studies Observational Epidemiology often key here….
Allows capitalization on “natural” or Case - control studies “unplanned” experiments.
¾ Take advantage of groups who have been exposed Cohort studies for non-study purposes.
Randomized trials*
*(if potential beneficial intervention identified)
Correlation between dietary fat intake and breast cancer by USA Ecologic Study 250 country. Switzerland
Canada Units of analysis are populations or groups Fed. Repub. Units of analysis are populations or groups 200 Italy Of Germany Israel UK Denmark of people, rather than individuals. Sweden France New Zealand Australia 150 Norway Finland Often exploit pre-existing data collected for Yugoslavia Spain 100 Poland Romania other purposes Hong Kong Hungary ¾ ¾ Efficient and economical design 50
Incidence Ratio per 100,000 Women Japan
0
0600 800 1000 1200 1400 1600 Prentice RL, Kakar F, Hursting S, et al: Aspects of Per Capita Supply of Fat Calories the rationale for the Women’s Health Trial. J Natl Cancer Inst 80:802-814, 1988.)
3 Key potential limitation: The ecologic Cross-sectional Study fallacy Draw sample from population of interest at Attributing to members of a group characteristics particular time that they do not possess as individuals Identify cases and non-cases of disease E.g., only know average values of fat consumption Measure characteristics (exposures) by country Examine associations between characteristics ¾ Don’t know if individuals with breast cancer had Don’t know if individuals with breast cancer had and disease higher fat intake
Example: Is stress associated Stress and TMD with symptoms of TMD? Percent reporting frequent stress: Random sample of population (N=680)
Those with TMD symptoms = 56% Interviewed re: symptoms of TMD (pain, joint sounds, limited opening) Those without symptoms = 21%
Measure of life stress P<0.05
Cross-sectional studies: Cohort Study
Can assess associations
Exposed Not Exposed Cannot establish correct temporal relationship for inferring causation
Why? Do Not Do Not Develop Develop Develop Develop Disease Disease ¾ Factor and disease measured at same point Disease Disease in time
4 The Cohort Concept Following the cohort through time…
t0 t1 t2
Analytical Design of a Cohort Study …to the end of the study period.
Then follow to see whether… Incidence Rates of Disease Disease does Totals develops not develop Disease a Exposed a b a + b a + b
First select c Not exposed c d c + d c + d
End of Follow-Up
Relative Risk Calculation for Cohort Relative Risk Study tsts en group Develop d Do Not Incidence d e CHD DevelopTotal population per 1,000 CHD Totals per Year Risk in exposed Expos New inci Smoke 84 2,916 3,000 28.0 (= 84/3000) Risk in non-exposed Cigarettes ( x 1,000 ) Do not smoke 87 4,913 5,000 17.4 (= 87/5000) Cigarettes ( x 1,000 ) RR = 28.0 / 17.4 = 1.61
5 Advantages of cohort studies Interpreting RR of a Disease Temporal relationship more certain If RR = 1 Risk in exposed equal to risk in unexposed (no association) Less opportunity for distortion of If RR > 1 Risk in exposed greater than risk in exposure data unexposed (positive association; Can examine multiple disease possibly causal) outcomes If RR < 1 Risk in exposed less than risk in unexposed (negative association; possibly protective)
Drop out in cohort study of oral Disadvantages of cohort studies health of older adults
Can be time consuming and expensive Baseline N=907 – follow large group over long periods of time 3 years N=611 Potential bias due to drop outs from study 7 years N=425
Concurrent Cohort Study Begun in 1995 Concurrent The Two Major Flavors of 1995 Defined Population Non-Randomized (Occurs naturally) Cohort Studies: It’s All in the
Timing 2005 Exposed Not Exposed
Do Not Do Not Develop Develop 2015 Develop Develop Disease Disease Disease Disease
6 Retrospective Cohort Study Begun in 1995
Retrospective Defined Population 1975 Non-Randomized
Exposed Not Exposed 1985
Do Not Do Not Develop Develop Develop Develop 1995 Disease Disease Disease Disease
7