Disease Detectives BIG IDEAS + EQUATIONS

Classical Epidemiology

Classical Epidemiology

studies origins of health problems

Clinical Epidemiology

study patients in health care settings in order to improve the diagnosis and treatment

Incidence

# of new instances of disease in a population over a given time period

Measures rapidity of disease occurrence

X cases / Y population / Z time

Time = the duration of the illness or condition

Prevalence

# of affected persons in the population at any given point in time

Measures the proportion of the population with disease \n \n X cases / Y population

Point Prevalence

snapshot of the population + its rate of a certain disease at a point in time

Period Prevalence

tracks the prevalence over a certain duration

13 Steps to Investigate Outbreak

1. Prepare for field work - Research and Supplies, Official Arrangements, Safety Protocols, and Contacts

1. Establish the Existence of an Outbreak - Consider Severity, Potential for Spread, Public Concern, and Availability of Resources

1. Verify the Diagnosis - Verify Procedures and Eliminate Experimental Error (and Other Errors/Biases, for That Matter)

1. Construct a Working Case Definition

1. Find Cases Systematically and Record Information - Time: Tables, Epi Curves; Place: Geographical Extent of Disease & Spot Map; Identify By Demographic Information or Exposures to Risk Factor

1. Describe and Orient the Data in Terms of Person, Place, and Time - Descriptive Epidemiology

1. Develop Hypotheses (Agent/Host/Environment Triad) = Chain of Transmission

1. Evaluate Hypotheses - Analytical Studies (MUST Have a Control Group)

1. Refine Hypotheses if Necessary

1. Compare and Reconcile with Laboratory and/or Environmental Studies

1. Implement Control and Prevention Measures (ASAP!)

1. Initiate or Maintain Surveillance - Monitor Implementation: Track New Cases, Check the Outbreak’s Spread Outside Targeted Area, Control and Change if Needed

1. Communicate Findings - Reports, To Important People and Public

5 Steps for Surveillance

1. Identify, define, and measure the health problem of interest

1. Collect and compile data about the problem (and if possible, factors that influence it)

1. Analyze and interpret these data

1. Provide these data and their interpretation to those responsible for controlling the health problem

1. Monitor and periodically evaluate the usefulness and quality of surveillance to improve it for future use. (Surveillance of a problem often does not include actions to control the problem

Odds Ratio

the odds of an event in an experimental group relative to that in a control group

Relative Risk

ratio of risk of an event in one group (e.g., exposed group) versus the risk of the event in the other group (e.g., nonexposed group)

Attack Rate

rate that a group experienced an outcome or illness equal to the number sick divided by the total in that group

For exposed: a/(a+b) → should be HIGH

For unexposed: c/(c+d) → should be LOW

Chi-Square

used to determine the statistical significance of the difference indicated by the relative risk or odds ratio; compares your observed values (a, b, c, and d) with the expected values for those same groups

Expected Value: [ (a+b)(b+c) ] / (a+b+c+d)

P-value

the measure of how confident you are that your findings are NOT due to chance

If P-value is LESS than alpha (0.05 or 5%), the data is significant

ex) P-value of 0.01 → 10% chance your results were a result of random fluctuations

Epi-Curves

a histogram (graph consisting rectangles) that shows the course of an outbreak by plotting the number of cases of a condition according to the time of onset

Point Source

persons are exposed over a brief time to the same source, such as a single meal or an event

Shape of curve (commonly) RISES RAPIDLY + contains a definite peak, followed by a gradual decline

Continuous Common Source

when persons are exposed to the same source but exposure is prolonged over a period of days, weeks, or longer

*the down slops of the curve may be very SHARPif common source is removed or gradual if the outbreak is allowed to exhaust itself

Propagated (Progressive Source)

when one or more of the first wave of cases serves as a source of infection for subsequent cases and those subsequent cases, in turn, serve as sources for later cases

SHAPE: curve usually contains a series of successively larger peaks (reflects increasing # of cases caused by person-to-person contact)

distance between peaks may be rough indication of the incubation period of the disease

As outbreak progresses, peak flattens out

Random Error

the result of fluctuations around a true value because of sample population; use of invalid outcome. measure that equally misclassifies cases. and. controls

can be reduced by: INCREASING sample size + making measurements more precise

Precision

measure of random error that is inversely related (INCREASING random error DECREASES precision)

Systematic Error

usually consistent and repeatable and often occurs from flawed equipment or experimental design

Selection Bias

when selection of participants for a study is affected by an unknown variable that is associated with the exposure and outcome being measured

Information Bias

when bias (any systematic error) is introduced though an error in measurement or observation

Confounding bias

results from mixing effects of several factors (deals with causation and NOT variations in study results)

Hill’s Criteria for Causation

nine viewpoints by which to evaluate human epidemiologic evidence to determine if causation can be deduced

1. Strength of Association - relationship is clear and risk estimate is high

2. Consistency - observation of association must be repeatable in different populations at different times

3. Specificity - a single cause produces a specific effect

4. Alternative Explanations - consideration of multiple hypotheses before making conclusions about whether an association is causal or not

5. Temporality - cause/exposure must precede the effect/outcome

6. Dose-Response Relationship - an increasing amount of exposure increases the risk

7. Biological Plausibility - the association agrees with currently accepted understanding of biological and pathological processes more exposure = more risk

8. Experimental Evidence - the condition can be altered, either prevented or accelerated, by an appropriate experimental process

9. Coherence - the association should be compatible with existing theory and knowledge, including knowledge of past cases and epidemiological studies

Infant Mortality Rates

ratio of deaths to births

Sensitivity

the chance of testing positive if you do have the disease

Specificity

the chance of testing negative if you do not have the disease

Attack Rate

total number of new cases / total population

Case Definitions

a set of standard criteria for classifying whether a person has a particular disease, syndrome, or other health condition

includes criteria for person, place, time, and clinical features

Length Time Bias

overestimation of survival duration due to the relative excess of cases detected that are asymptomatically progressing, while fast progressing cases are detected after giving symptoms

Simpson’s Paradox

Women appeared to be more susceptible (influenced) to an illness than men, but when studies were carried out, men were found to have higher probability of contracting illness

Pigmalion Effect

researchers convey high expectations to subjects; the subjects produce those results

Cross Sectional

a survey, health questionnaire, “snapshot in time”

Fastest

Least Expensive

Good for more than 1 Outcome