Surveillance and Epidemiological Investigation (22 questions) Flashcards
Quantitative
Generates numerical data, represents counts or values. Seeks to establish casual relationships between two or more variables- uses statistic methods to test strength and significance
Cross-sectional study
Survey of INDIVIDUALS that assesses both exposure and disease at the same time.
examine outcome and risk factors in a population group at one point in time. Provide “snapshot” (Prevalence, correlation or survey studies)
Case-control study
assemble group of individuals with disease as well as a comparative group without disease; then investigate proportions with the exposure of interest in each group.
examine population of individuals with and without an OUTCOME of interest, studied for exposure to one or more risk factors.
Quicker, less expensive and easier
Calculate odds ratio
Ex: Wanting to figure out source of outbreak
Cohort study
assemble a group of individuals with an EXPOSURE and a comparative group without. Ensure all individuals are free of disease at start of the study, then follow this group over time investigating the proportion that subsequently develop disease in each group.
same of individuals with and without EXPOSURE to potential risk factor who are followed for incidence of outcome in each group
less pt selection and stronger evidence of casual association
Calculate Relative risk/risk ratio
Qualitative
based on description and observation. for theory verification
Rate
x/ y x k
x= numerator (# times event occurred)
y = denominator, population (ex. # pt at risk)
k = constant used to transform results of division into uniform quality
Incidence rate
a measure of the frequency with which new cases or events occur among a population during a specified period.
NEW cases/(pop at risk x constant)
ex: new cases/ total patient days x constant
Attack Rate
new cases/#exposed to risk factor
Type of incidence rate
(# new cases/pop at risk) x100
new cases/ # people exposed
Prevalence Rate
of EXISTING cases/ pop at risk x 100
Mortality rate
(# of deaths/pop at risk) x 10,000
crude= all causes of death
cause specific = rate from a certain disease
SIR
observed/predicted
Sensitivity
% of true positive
a/(a+c) x100
TP/ (TP+FN)
true positives/ # with outcome x 100%
If someone has the outcome, what is the likelihood the test will be positive?
Specificity
true negatives/# individuals without outcome x 100%
% of true negatives
d/ (b+d) x 100
TN/ (TN+FP)
If someone does not have the outcome, what is the likelihood the test will be negative?
Positive predictive value
true positives/ # individuals with positive result (TP+FP) x 100%
If the test result is positive, what is the likelihood that the person truly has the outcome?
Negative predictive value
true negatives/ # with neg result x 100%
If the test result is negative, what is the likelihood that the person truly does not have the outcome?
Validity
the degree to which a screening test or other data collection tool measures what it is intended to measure.
P value
between 0 to 1:
helps determine significance of results
- small P value (≤ 0.05) strong evidence against the null, reject null hypothesis. statistically significant.
- Large p value (>0.05) weak evidence against null hypothesis, fail to reject null. Not enough evidence to suggest null is false.
Power
probably that you will reject the null hypothesis when you should
The power of a test is its ability to detect a specified difference
(e.g., the probability of rejecting the null hypothesis when it is false). The
power of a hypothesis test is affected by three factors:
1. Sample size (n). In general, the greater the sample size, the greater
the power of the test.
2. Significance level (α). The higher the significance level, the higher
the power of the test.
3. The “true” value of the parameter being tested. The greater the
difference between the “true” value of a parameter and the value
specified in the null hypothesis, the greater the power of the test.
That is, the greater the effect size, the greater the power of the test.
Confidence interval
estimated range of values likely to include an unknown pop parameter.
width gives an idea of how uncertain. Wide interval = more data needed
usually collected at 95%
standard deviation
variability in values around the mean. Distance between data point and mean.
- positive deviation= greater than mean
- negative deviation= less than mean
- no deviation= equals mean
small= tight grouped, precise
large= spread out
emperical rule:
68% = 1 standard deviation from mean
95%= 2 standard deviation from mean
99.7= 3 standard deviation from mean
Data normally distributed
Mean, Median and Mode are equal, bell shape curve
? Deep SSI, 90 days
Craniotomy, coronary artery bypass graft
Process measure
focuses on process that lead to specific outcome. Process measures are commonly used to evaluate
compliance with desired care or support practices or to monitor variation in
these practices (Ex: app abx usage, med errors, flu vaccination rates, sterilization quality assurance testing, compliance with iso protocols, TB skin testing, HH, environmental cleaning, communicable disease reporting, antimicrobial prescribing and administrations, maintaining barriers during constuction)
Outcome measure
indicates the result of the performance (or non-performance) of a function or process (ex: HAI’s, infection or colonization of specific org, flu or TB skin test conversions, sharps injuries, decubitus ulcers, pt falls, CLABSI)
Attributable risk percent
gives the proportion of cases attributable (and avoidable) to this exposure in relation to all cases. calculate in order to figure out diseases that could be prevented by eliminating exposure in the entire study population
(relative risk - 1) / relative risk
precision of relative risk is related to
the power of the study. Power is affected by size of the effect and the size of study/sample size
most common reservoir for highly pathogenic avian flu H5N1 virus
birds, esp domestic poultry (chicken, turkeys, geese, ducks)
Casual Web
causation is essential concept in epidemiology. The web of causation refers to the interrelationship of multiple factors that contribute to the occurrence of disease
indirect and direct causes of disease may form a complex network of events that determines the level of disease in a community
Higher prevalence of disease
will increase the PPV (positive predictive value) and decrease the NPV (neg predictive value)
Most common cause of epidemic gastroenteritis worldwide and leading cause of foodborne outbreaks in US
Norovirus
Attributable risk
Difference in rate of a condition between an exposed pop and unexposed
AR= IE - IU
IE (Incidence of exposed) = # exposed who get disease / total # exposed
IU (unexposed) = # unexposed with get disease / total # unexposed
Odds ratio
probability of having a particular risk factor if a condition/disease is present divided by the probability of having the condition/disease is not present
ad / bc
2x2 table:
A| B
—–
C| D
(p = 0.001) p value indicates
more evidence in support of which
The alternative hypothesis (reject null and accept alternative hypothesis)
Rationale: The p value is the probability of obtaining the observed sample
results (or a more extreme result) when the null hypothesis is actually true.
If the p value is small (≤ the significance level), it suggests that the observed
data is inconsistent with the assumption that the null hypothesis is true, and
thus that hypothesis must be rejected and the alternative hypothesis accepted
as true.
Risk ratio or relative risk
Used in cohort studies
% Risk of outcome in group WITH exposure/ risk of outcome of group WITHOUT exposure
Incidence proportion= cases/ exposed or unexposed. Use this to calculate % of With and without exposure
Type I error
rejecting the null hypothesis when it is true
p value = probability of type 1 error
A Type I error occurs when one rejects the null hypothesis (H0)
when it is true. This is also called a false-positive result (as we incorrectly
conclude that the research hypothesis is true when in fact it is not). The p
value or calculated probability is the estimated probability of rejecting the
null hypothesis of a study question when that hypothesis is true.
Type II error
failing to reject a false null hypothesis
What information will be needed to calculate a
CLABSI rate for the ICU?
BSI and # device days
Need device days, NOT # central line
The number of patients who had bloodstream infections
identified and the number of device days for the time period
X/Y x K
#BSI/device days x K
superficial SSI
occur within 30 days-only skin or subcutaneous tissue.
In addition, one of the following must be met:
*Purulent drainage,
*Organisms isolated from an aseptically obtained culture of fluid or tissue
And patient has at least one of the following:
°° Purulent drainage from the superficial incision
°° Organisms isolated from an aseptically obtained culture of fluid or
tissue from the superficial incision
°° Superficial incision that is deliberately opened by a surgeon,
attending physician, or other designee*
And patient one signs or symptoms: pain
or tenderness, localized swelling, redness, or heat. A culture negative
finding does not meet this criterion.
°° Diagnosis of superficial incisional SSI by the surgeon or attending
physician or other designee
Chi-square tests (χ2)
-To evaluate the effect of a variable on outcomes
-To calculate an odds ratio or relative risk
- If each cell of the table is greater than 5
Chi-square tests (χ2) can be used to test the association between
two classifications of a set of counts or frequencies (discrete data). This data
are commonly displayed as a contingency table or 2 x 2 table where rows
represent one variable and columns represent the other. The null hypothesis
is that there is no association between the two variables. Row and column
totals (marginal totals) are used to predict what count would be expected for
each cell if the null hypothesis were true. A test statistic is calculated from the
observed and expected frequencies. The larger the test statistic (for given
degrees of freedom) the more likely there is to be a statistically significant
association between the two variables. Chi-square tests are used for medium
to large samples (see Figure 4-1). The Fisher’s exact test is used in place of the
χ2 when the sample size number is less than 20 or any one cell in the table is
less than 5.
Figure 4-1. Formula for chi-square
χ2 =
(0-E)2 /
E
Where:
O = observed frequency
E = expected frequency
The measure of central tendency most affected by outliers is
Mean
Which statistical test is used when the data are small in numbers?
Fisher’s exact
Rationale: Fisher’s exact test is a statistical significance test used in the
analysis of contingency tables. Although in practice it is employed when
sample sizes are small, it is valid for all sample sizes.
A measure of dispersion that reflects the variability in values around the mean is called the:
Standard deviation
Steps to hypothesis testing include
- State the research question
- Specify the null and alternative hypotheses
- Calculate test statistic
- Compute probability of test statistic or rejection region
- State conclusions
Analysis and dealing with outliers is an important component of statistical analysis. Sometimes careful analysis of outliers, their removal, or weighting down
can change the conclusions considerably. Outliers should be investigated to
determine the optimal method of analysis.
If the index of kurtosis is –1.99, then the curve is
Relatively flat
Rationale: Two terms are used to describe the shape of a frequency
distribution: “skewness” and “kurtosis.”
Kurtosis refers to how flat or peaked a curve is (see Figure 4-3):
* Leptokurtic is the more peaked curve. (+) positive numbers
* Mesokurtic is a typical bell-shaped curve or normal distribution. A value of 0
* Platykurtic is the flatter curve (-) negative numbers.
The term for an extraneous variable that systematically varies
with the independent variable and influences the dependent
variable is a
Confounding variable
Rationale: A confounding variable is a variable that has an important
confounding effect on the result but is not among the variables being
studied. It can suggest a false relationship between variables, or it can
hide a relationship that exists.
frequency polygon
useful for showing 2 sets of data on a graph, uses connecting lines and data points
virus is the causative agent in
Kaposi’s sarcoma?
Human herpesvirus 8
when the prevalence
of a disease is very low
The positive predictive value of a diagnostic test is lowered
Sensitivity and specificity are independent of prevalence of disease. PPV and
NPV are disease prevalence dependent. Generally a higher prevalence will
increase the PPV and decrease the NPV.
Norovirus
Noroviruses (NoVs) are the most common cause of epidemic gastroenteritis worldwide and the leading cause of foodborne outbreaks in
the United States. Severe disease associated with NoV occurs most frequently among older adults, young children, and immunocompromised patients. NoV outbreaks occur year round, but activity increases in the United States duringthe winter months; 80 percent of reported outbreaks occur during Novem ber–April. Most NoV outbreaks are attributed to genotype GII.4, which evolves
rapidly over time.
point source epidemics
exposed to same source over brief period of time (ex: meal or event)- transmission occurs through a vehicle (same person) rather than person-to- person
criteria for causality
*The incidence of disease is higher in those who are exposed to the factor
*Evidence that the independent and dependent variables are related
*The association has been observed in numerous studies
Rationale: The criteria for causality are known as Hill’s criteria and use
epidemiological methods to determine whether a factor is causal for a given
disease. Hill’s criteria for causation are as follows:
1. Strength of association: The incidence of disease should be higher in
those who are exposed to the factor under consideration than in those
who are not exposed; that is, the stronger the association between an
exposure and a disease, the more likely the exposure is to be causal. For
example, lung cancer is common in those who smoke.
2. Consistency: This means that the association should be observed in
numerous studies, preferably by different researchers using different
research methodologies.
3. Specificity: Refers to an association between one factor and one
disease, and this association is more likely to be causal. This criterion
also refers to the extent to which the occurrence of one factor can be
used to predict the occurrence of another (disease). In reality, such a
one-to-one relationship is rare due to the multifactorial causes of most
diseases and because, sometimes, the same factor(s) can cause more
than one disease.
4. Temporality: This must also be addressed when determining cause of
disease. Essentially, exposure to the hypothesized causal factor must
precede the onset of disease.
5. Biological gradient: The biological gradient is a dose-response
relationship between increased exposure to a factor and increased
likelihood of disease. For example, the longer one smokes, the more
likely one is to develop lung cancer. If the association demonstrates a
biological gradient between the factor (exposure) and effect (disease),
the relationship is more likely to be causal.
6. Plausibility: The association in question should also be biologically
plausible in light of current knowledge. This criterion may be the most
elusive and variable of the nine. Because biological knowledge is ever
expanding, lack of biological plausibility does not necessarily disprove
a theoretical association.
7. Coherence: There should be coherence between known information
about the biological spectrum of the disease and the associated factor,
that is, the association should be in accordance with other facts known
about the natural history of the disease.
8. Analogy: Associations derived from experiments add considerable
weight to evidence supporting causal associations. These experiments
can be animal model studies or clinical trials; however, although animal
models may be helpful, many diseases do not manifest the same way in
animals and humans.
9. Finally, if similar associations have been shown to be causal, by analogy
the association is more likely to be causal. Determining causality may
also help to determine at which points the natural history of a disease
may be interrupted, so that prevention and control efforts are effective.
It can also add information on the natural history of a disease.
scientific criteria for disease causation
Koch’s postulates consist of four points:
- The organism must always be found with the disease, in accordance with the clinical stage observed.
- The organism must then be grown in pure culture from a diseased host.
- The same disease must be reproduced when a pure culture of the organism is inoculated into a healthy susceptible host.
- The organism must then be recovered from the experimentally infected host.
A causal association should not be confused with causality, which requires a number of conditions to be met, one of which is the presence of causal associations
