microbio: epidemiology and public health

this lecture introduced some basic concepts about epidemiology and public health. epidemiology is the study of how, why, where, when diseases are transmitted within populations and public health is concerned with preventing the transmission of disease within populations. some common measures that are used as tools in epidemiology are incidence and prevalence- how many new cases of a disease there are compared to the people at risk, and the number of old and new cases of a disease compared to the people at risk, respectively. incidence rate looks at the incidence divided by "people-years", which is a measure of the summed up at-risk years of an entire population. average risk or incidence proportion is another measure which is used and is the cases of disease occurrence divided by the population size. finally, "incidence odds" is the number of people with a certain outcome vs. the number who don't get the outcome.

there are different types of epidemiology which use different approaches to study disease transmission in populations. descriptive epidemiology makes use of descriptive factors such as location, time, patient characteristics in order to investigate time courses of disease and chains of transmission. the most prominent (and one of the first) example is John Snow in 1854 discovering that a cholera outbreak was traced back to a single well in Soho England. analytical epidemiology makes use of descriptive data as well, and also looks at retrospective studies to establish etiology of disease transmission (and thus can sometimes be used in place of Koch's postulates). experimental epidemiology forms hypotheses based on descriptive or analytical data and performs tests involving treatment or intervention (and thus can sometimes have ethical issues). lastly, hospital epidemiology deals with nosocomial infections exclusively, including the iatrogenic cases in which disease occurs directly due to treatment, and from cases in which the disease is transmitted from patient to patient.

diagnostic testing is an important facet of epidemiology; as it is the method of acquiring data, it is vitally important to scrutinize as well. the quality of a particular diagnostic test in assessing disease is gauged by the two variables: validity (whether the test actually yields accurate results) and reliability (whether these results can be repeated). sensitivity of a test is defined as the percentage of people with the disease that test positive for that disease. specificity is the percentage of people without the disease that test negative for the disease. conversely, a positive predictive value is the percentage of people who have positive test results that actually have the disease and negative predictive value is the percentage of people who test negative and don't have the disease.

a brief discussion about error and bias in collecting epidemiological (or any) data. while random error is due to statistical aberrations that are expected in small sample sizes, systematic error is due to a fundamental flaw in the mechanism for collecting data and thus can not be eliminated by large sample sizes. confounding bias is when two populations that are being compared have some innate difference between them that does not for a fair comparison. selection bias is the error introduced by a study when the methods for selecting participants of a study favor certain groups. information bias is a misclassification of the study participants with respect to disease or exposure status.

questions
introduction...
1. define epidemiology.
2. epidemiological factors were originally limited to infectious diseases but now include...
3. public health is concerned with...
4. all epidemiological studies should have...

incidence and prevalence...
5. define "incidence".
6. what is the "incidence rate"?
7. what is the second most common measure of disease occurrence?
8. what is the third most common measure of disease occurrence?
9. what type of data does an incidence case control study use?
10. define "prevalence".
11. define the following descriptions of prevalence: endemic, sporadic, epidemic, pandemic.

different types of epidemiology...
12. what characteristics of disease occurrence does descriptive epidemiology make use of? what is a prominent example of descriptive epidemiology at work in uncovering a cholera outbreak?
13. describe analytical epidemiology.
14. describe experimental epidemiology.
15. describe hospital epidemiology.

diagnostic testing...
16. what are two abstract measures that gauge the quality of a particular diagnostic test?
17. define "sensitivity" in terms of the quality of a particular diagnostic test.
18. ...specificity.
19. ...positive predictive value.
20. ...negative predictive value.
21. what are some characteristics of a disease that is worthwhile for diagnostic screening?

22. what is the difference between random error and systematic error?
23. what is confounding bias?
24. what is selection bias?
25. what is information bias?

answers
1. the study of how, where and when diseases occur within populations.
2. car accidents, cigarette smoking, lead poisoning, injury, etc.
3. prevention of disease in populations.
4. source or study population and risk period.

5. number of new cases of a disease / number of people at risk for that disease.
6. disease occurrence / number of "person-years" (the population size times years at risk)
7. average risk or incidence proportion: cases of disease occurrence / population size.
8. incidence odds: number who get outcome / number who don't get outcome.
9. retrospective data, which can be less expensive and resource heavy than a regular prospective study.
10. number of old cases + number of new cases / number of people at risk.
11. endemic: when disease occurs continually or regularly. sporadic: a few scattered cases. epidemic: greater frequency than endemic, more cases than historical statistics indicate. pandemic: simultaneous epidemic on more than 1 continent.

12. location, time, patient characteristics; looks at time course of disease and investigates chains of transmission. prominent example is John Snow, who in 1854 helped trace the source of a cholera outbreak to contaminated water from a single well in Soho, England by talking to local residents and analyzing statistics from the water delivery company.
13. makes use of descriptive data and retrospective studies. may be used in place of koch's postulates in establishing etiology.
14. hypothesizes based on descriptive or analytical data and performs tests (can be treatment or intervention based).
15. studies nosocomial infections exclusively. exogenous infection is derived from an agent outside of patient. endogenous infection comes from patient. iatrogenic infection is the direct result of medical procedure.

16. validity and reliability.
17. percentage of people with disease with positive test results for that disease.
18. percentage of people without disease with negative test results for that disease.
19. percentage of people with positive test results who have the disease.
20. percentage of people with negative test results who don't have the disease.
21. a disease with a high burden of human suffering, effective early intervention, a high risk population, has an effective and feasible diagnostic method.

22. random error is the natural deviation that occurs from predicted values which is statistically expected with low sample sizes, whereas systematic error's basis is a lack of validity in the tools of discernment used in the study and thus cannot be corrected for with larger sample sizes.
23. a bias which affects studies of cause and effect; when exposed and non-exposed groups are not comparable due to inherent diferrences in background or risk factors.
24. a bias that arises because of the particular way in which participants in a study are selected.
25. a misclassification of the study participants with respect to disease or exposure status.