Term
| 2 main divisions of epidemiology |
|
Definition
causal (etiology)
Descriptive (diagnostic, prognostic, intervention) |
|
|
Term
| What did the Ioannidis, JP; JAMA 2005 that reviewed 45 high profile studies in NEJM, JAMA, and Lancet find? |
|
Definition
|
|
Term
| Authoritative v. experimental research |
|
Definition
authoritative: textbooks, professors, experts, specialists, colleagues
experimental: evidence-based medicine |
|
|
Term
| Components of a "good" study |
|
Definition
unbiased
good control of variables, placebo effects, and observer bias
large sample size |
|
|
Term
| the gold standard study design in clinical research |
|
Definition
| double blind, placebo controlled |
|
|
Term
| What is the purpose of blinding a study? |
|
Definition
patient: to control placebo effect
researcher: control observer bias |
|
|
Term
| a multidisciplinary field whose goal is to promote the health of the population through organized community efforts |
|
Definition
|
|
Term
| how did the US life expectancy rise from 47 years in 1900 to 77 years in 2000? |
|
Definition
mostly due to public health achievements: sanitation, quality of water, less food poisoning, less toxic exposure, reduced smoking, preventative medicine
also due to medical care improvements (medications) |
|
|
Term
| the study of that which falls on the common people |
|
Definition
|
|
Term
| the study of the DISTRIBUTION & DETERMINANTS of disease frequency in human populations and the APPLICATION of this study to control health problems |
|
Definition
|
|
Term
1. studied mortality rates in london circa 1662 - what did he find?
2. performed a controlled study of pts with scurvy at sea by treating 12 pts w/ 6 different treatments. what did he find?
3. the "father of epidemiology" - demonstrated a link between cholera and contaminated water in 1800s |
|
Definition
1. john graunt
lots of deaths in early childhood
2. James Lind - citrus fruits did best
3. John Snow |
|
|
Term
| how have the leading causes of death in the US changed over the past 100 years? |
|
Definition
1900 - infectious diseases led
2000 - chronic diseases (Heart disease, cancer, stroke, COPD) led.
% of accidental deaths haven't changed
cigarettes began being made around 1900
|
|
|
Term
| Most of the reason that death rates from infectious diseases have fallen is what? |
|
Definition
public health and sanitation -- NOT abx
|
|
|
Term
| New directions of epidemiology? |
|
Definition
genetic and molecular
social (overcrowding, city v/ rural) |
|
|
Term
|
Definition
1. certain infectious diseases
2. rare presentaitons of other diseases - 15 y/o who dies of vaginal cancer |
|
|
Term
| Purpose of an introduction |
|
Definition
quick review - bring reader up to date on general topic
elucidate particular issue in study
succinctly state study design |
|
|
Term
|
Definition
sponsored by US public health service
400 black men w/ syphilis
no informed consent
misinformation: tests were presented as tx
penicillin denied to subjects
wives and unborn kids were infected
Went on for 30+ years
|
|
|
Term
nuremberg code
1. ____ consent of human subjects is essential
2. ____ experimentation should precede human experimentation
3. all unnecessary ___ & ___ should be avoided
4. the degree of risk to the participants should never exceed the _______.
5. participants should always be able to ____ the experiements w/o penalty |
|
Definition
dervied from 1946 nuremberg doctors trial
1.voluntary
2. animal
3. suffering and injury
4. humanitarian importance of the problem
5. withdraw from
|
|
|
Term
| brooklyn jewish chronic disease hospital |
|
Definition
1956 (after nuremberg) study to determine if body's failure to reject cancer cells was due to the cancer itself or to debilitation
pts injected w/ live cancer cells w/o being told |
|
|
Term
statment by the world medical association in 1964
- attempt to improve on nuremberg code
- addressed medical research w/ therapeutic intent
- firt time international medical commmunity attempted to polic itself
- journal editors required that research be performed in accordance with this
still lacked force of law
key elements:
1. research must include ____.
2. human subjects should be reviewed by an ____
3. human research should be preceded by ___
4. risks should not outweigh possible ____ |
|
Definition
declaration of helsinki
1. freely given informed consent
2. independent committee
3. lab reserach
4. benefits |
|
|
Term
|
Definition
required independent committee review of human subject research
= IRB
became law in 1974 - US national research act |
|
|
Term
the belmont report:
- product of the national commission for the protection of human subjects of biomedical and behavioral research
- published 1979
3 major principles: |
|
Definition
respect for persons
beneficence - minimize risk and maximize benefits. personal info must be safeguarded
justice
|
|
|
Term
| elements of informed consent (5) |
|
Definition
explains how research is conducted
describes the benefits that may be reasonably expected
describes alternative tx
describes possible tx related events
describes legal rights to bail at any time |
|
|
Term
| 1974 national research act - describes federal legal protections for ? |
|
Definition
|
|
Term
|
Definition
protection of human research subjects
required before any study w/ human subjects
not done to evaluate the science, just the ethics and pt protection
1. appropriate informed consent
2. adequate protection of privacy
3. miminization of risks
4. risks reasonable compared to benefits
5. special levels of protection for vulnerable subjects
6. fairness in subject selection |
|
|
Term
| can the results of a study be withheld if the results are unfavorable? |
|
Definition
| no - have to report to government |
|
|
Term
| a condition w/ signs and/or sx that is linked to an increase risk of future death or disability |
|
Definition
|
|
Term
|
Definition
transmissible
non-transmissible |
|
|
Term
| examples of cultural transmission of disease |
|
Definition
drug use
obesity
occupational
poverty |
|
|
Term
| impact of poverty on disease |
|
Definition
nutrition
living environment
immune status
health care |
|
|
Term
1. looks at frequency of new cases (the fraction of a population that develops) of a disease which occurs per 100,000 people in a certain period of time
2. look at total number of cases present at any one moment in time per 100,000 people -- fraction of a population possessing a clinical condition or outcome at a given point in time |
|
Definition
1. incidence
2. prevalence |
|
|
Term
| first step in studying a disease |
|
Definition
|
|
Term
| what are some ways in which incidence is determined? |
|
Definition
1. reportable case data - STDs, TB
2. hospital discharge diagnoses
3. office outpt data
4. prospective cohort studies
5. can be calculated if disease duration and prevalence is known |
|
|
Term
| what are some ways in which disease prevalence is determined? |
|
Definition
1. measure/assess the population
2. measure/assess a sample of the population -- more common
3. calculate if incidence and duration are known |
|
|
Term
| calculating disease incidence and prevalence |
|
Definition
prevalence = incidence x disease duration
incidence = # who develop disease/total population
(12 newly dx DM in 1 year/out of a community of 10,000) |
|
|
Term
| when are disease prevalance and incidence the same? |
|
Definition
when disease processes are very short - ex: URI
never the same for chronic illness |
|
|
Term
| when calculating disease incidence/prevalence, what are some things that can influene the numerator and denominator? |
|
Definition
numerator: inaccurate diagnoses
denominator: gender, ethnicity, age |
|
|
Term
the study of the distribution and determinants of the disease frequency in human populations and the application of this study to control health problems
focuses on disease DISTRIBUTION
critical elements: person, place, and time
strengths and weaknesses? |
|
Definition
descriptive epidemiology
strengths: very good at generating disease causation hypotheses
weakeness: almost never able to prove disease causation conclusively |
|
|
Term
|
Definition
people who are employed tend to have lower incidence of diseases
unhealthy people are less likely to be hired
|
|
|
Term
|
Definition
observational
- observe what happens to a group over a period of time
experimental
- involve an intervention (tx, exposure to risks) |
|
|
Term
| pros and cons to experimental studies |
|
Definition
pros: more scientifcally rigorous, better control of variables
cons: more difficult/expensive, may be prohibited by ethical isues |
|
|
Term
| 4 types of observational studies |
|
Definition
cohort
case control
cross-sectional
ecological |
|
|
Term
2 types of cohort observational studies
advantages & disadvantages of cohort studies? |
|
Definition
prospective - observe for disease outcome at future points of time
- ex: framingham
retrospective
- determine disease incidence and risk exposure now
advantages: establishes incidence directly, logical, exposure can be elicited w/o bias of already knowing the outcome, can assess relationship between exposure and a number of outcomes
disadvantages: inefficient (need a large number), expensive, results not available for many years, may be limited in that only risk recorded at outset can be followed |
|
|
Term
type of obervational study that starts with a group that already has a dz or conditions (cases) and match them with another group that doesn't have the disease (controls). compare risk factors exposures between two groups
advantages and disadvantages |
|
Definition
case-control
advantages: can be done quickly, results available quickly, costs are usually minimal
disadvantages: often difficult to match two groups in all significant parameters, potential bias of retrospective analyses, limited data regarding temporal association of risk exposure |
|
|
Term
type of observational study that evaluates the relationship between risk factors and disease in a population at one specific point in time
- provides a "snapshot" analysis of a specified population
- frequently used for public health planning and etiologic research |
|
Definition
|
|
Term
ex of what kind of study?
National Survey of Family Growth
Survey of 11,000 women aged 15-44
Pregnancy history, contraceptive use, breastfeeding practices, etc
Is breastfeeding protective against breast cancer? |
|
Definition
|
|
Term
evaluates rates of disease in a population rather than individuals
- studies the effects of potential risk factors on the population as a whole |
|
Definition
|
|
Term
ex of what kind of study?
effects of automobile pollution on the health of various metro areas |
|
Definition
|
|
Term
what kind of study?
1. Compare 100 patient with brain cancer to…
100 individuals without brain cancer
Compare cell phone use
2. Compare brain cancer incidence in a metropolitan area before cell phones were used to…
A time after widespread utilization of cell phones
3. Assemble a large cohort and follow for 5+ years
Compare cell phone use to cancer incidence
4. Assemble large cohort
Divide randomly into groups
Control (no phone use)
Limited cell phone use
Heavy cell phone use
Follow for 5+ years and determine brain cancer incidence |
|
Definition
1. case-control
2. cross-sectional
3. cohort study
4. experimental |
|
|
Term
| study results are presumed to be valid if what 3 alternative explanations are excluded? |
|
Definition
1. bias
2. confounding effects - distort the relationship btwn disease and an exposure
3. random error |
|
|
Term
1. systematic error in the design or conduct of a study that leads to an erroneous association btwn exposure and disease
2 main types?
- Error due to systematic differences in characteristics between those who partake in a study and those who do not
Error due to systematic differences in selecting and following the study groups. unlikely if participation rate is 80% or more for both groups
- Error due to systematic differences in the way information on exposure and disease is obtained in the study groups |
|
Definition
bias
selection or observational |
|
|
Term
| false polling results before the truman-dewey election in 1948 is an example of what kind of bias? |
|
Definition
| control group selection bias |
|
|
Term
What kind of observational biases?
1. different level of accuracy in information provided by the compared groups. most often seen in case-control studies. is known to occur but still incompletely understood. minimized by verifying information and only comparing diseased/affected groups.
2. systematic difference in soliciting, recording, or interpreting info. most problematic in studies w/ in-person or telephone interviews. can involve language/cultural issues.
limited by blinding the study to the interviewers
3. a mistake made in the classification of expsure or disease. differential and non-differential (tend to bias study result toward null hypothesis) |
|
Definition
1. recall bias
2. interviewer bias
3. misclassification (measurement error) |
|
|
Term
data which serves as a marker of overall health status
permits recognition of problems and issues, determining progress of health programs, appropriate focus of research efforts
common indicators: reproductive, mortality, population |
|
Definition
|
|
Term
| def of maternal mortality rate/ratio |
|
Definition
death of a woman while pregnant or within 6 weeks of delivery or pregnancy termination
highest in sub-saharan africa, lowest in ireland |
|
|
Term
| ____ is the gold standard in experimental studies of treatment efficacy |
|
Definition
| randomized controlled clinical trials |
|
|
Term
phases of clinical trials
pre-clinical
phase 1
phase 2
phase 3
phase 4 |
|
Definition
pre-clinical: lab tests in vitro and with animals, identify possible safety issues
phase I clinical trials: small group of normal subjects, dose ranging, pharmacodynamics & kinetics
phaseII: small groups of pts, usually RCCT. "proof of concept" study
phase III: large scale, usually multi-center RCCT
Phase IV: ongoing saftey and effectiveness surveillance after approval for general use |
|
|
Term
| do study patients change their behavior in some way that affects the outcome -- the ___ effect |
|
Definition
|
|
Term
| Type I and Type II errors |
|
Definition
Type I: treatments are not different but the study concludes they are
Type II: treatments are different but they are concluded not to be |
|
|
Term
___ determines if a tx works under ideal conditions
___ determines if a tx does more good than harm in those to whom it is offered under ordinary circumstances |
|
Definition
|
|
Term
| limitations of randomized trials |
|
Definition
- # of available participants w/ the desired criteria may be limited
- costs may be prohibitive
- outcomes may not be determined for many years |
|
|
Term
| true/false: it's possible to draw substantial conclusions from uncontrolled trials |
|
Definition
|
|
Term
a method to obtain pooled quantatitive or qualitative data analysis of the research literature on a particular topic
advantages: allows for aggregation of studies w/ appropriate measures, can increase statistical power due to cumulative sample size, can help give an overall perspective when studies disagree
can be subject to publication bias in favor of publishing statistically significant results |
|
Definition
|
|
Term
the application of epidemiology under a set of general conditions:
- problem is unexpected
- a timely response may be demanded
- travel to and work in the field is required by epidemiologists to solve the problem
- the investigation time is likely to be limited bc of the need for a timely intervention
steps:
1. establish the existence o fan epidemic
2. confirm the dx
3. establish criteria for case ID
4. search for missing cases
5. count cases
6. orient the data according to person, place, and time
7. classify the epidemic
8. determine who's at risk of becoming a case
9. analyze the data
10. formulate a hypothesis
11. test hypothesis
12. develop reports and inform those who need to know
13. execute control and prevention measures
14. administer and plan activities |
|
Definition
|
|
Term
components of field epidemiology:
1. describe rapidly occurring new cases of disease in a well-defined population over a limited time period calculated by person characteristics (age, sex, race, ethnicity, occupation)
2. an unusual aggregation, real or percieved, of health events that are group together in time and space and that are reported to a health agency |
|
Definition
1. attack rates
2. disease clusters |
|
|
Term
| components of an ideal screening test |
|
Definition
1. cheap, safe, easy
2. designed to detect a significant disorder for which early intervention will significantly improve prognosis
3. detects all those with the disease (no false negs)
4. does not produce pos results for anyone who does not have the disease (no false pos) |
|
|
Term
1. the ability of a test to correctly ID those with the disease
= the proportion of those w/ the disease who have a pos test
calculation?
2. the ability to correctly ID those w/o the disease
- the proportion of those w/o the disease who have a neg result
calculation?
3. proportion correctly identifed in terms of disease status w/ or w/o the disease
4. probability that an individual w/ a pos result actually has the disease
5. probability that an individual w/ a neg result truly does not have the disease |
|
Definition
1. sensitivity = true pos/(true pos+flase neg)
2.specificity = true negs/(false pos+true neg)
3. overall accuracy = (true pos+true neg)/total # tested
4. PPV = true pos/(true pos + false pos)
5. NPV = true neg/(true neg+ false neg) |
|
|
Term
| age, gender, and clinical factors which determine overall risk for the population being tested -- required for accurate predictive values |
|
Definition
|
|
Term
2030 patients tested for fecal occult blood, then undergo colonoscopy
30 are found to have colon cancer
20 had positive tests for fecal occult blood (TP)
10 were negative for fecal occult blood (FN)
2000 are found to not have colon cancer
180 had positive tests for fecal occult blood (FP)
1820 had negative tests for fecal occult blood (TN)
calculate sensitivity, specificity, overall accuracy, PPV, and NPV |
|
Definition
Sensitivity: 20/30 = 66.67%
Specificity: 1820/2000 = 91%
Overall accuracy: 1840/2030 = 90.6%
PPV: 20/20+180 = 10%
NPV: 1820/1820+10 = 99.5% |
|
|
Term
1. a false association between exposure and disease due to chance
2. an uncontrollable force that seems to have no assignable cuase - three "uns" - unsystematic, unforseeable, unpredictable
3. a probability calculation that the null hypothesis is correct - if high, null hypothesis is likely to be true
4. quantifies random error by quantifying the variability around the point estimate of association (risk), the wider, the more random error |
|
Definition
random error
chance
3. P value
- P of <0.05 (5%) is acceptable
4. confindence interval - 95% CI = 95% chance that the true measure of association lies somewhere in the CI |
|
|
Term
if CI is above risk factor of ___, then there is a significant Pos association of the exposure to the disease
if the CI is below __, then there is a significant neg association
if the CI is ___, there is not a significant association |
|
Definition
|
|
Term
A cohort study compares the incidence of peptic ulcer disease between two groups
The rate is 7.6 % among those taking drug X vs. 3.8% among those not taking the drug
Is this difference important/real)? |
|
Definition
•Relative risk is 2 but 95% confidence interval is 0.8 – 2.6
•Conclusion:
•Drug X does not result in a statistically significant increase in risk of peptic ulcer
|
|
|
