Practice Questions 4 Flashcards
What is lifetime prevalence?
Proportion of individuals in population who have ever manifested a disorder who are alive on a given day.
What is lifetime morbid risk?
Probability of a person developing a disorder using entire period their life
Which lifetime calculation includes those deceased?
Lifetime morbid risk
For which type of disorders does age-specific incidence rates give approximate lifetime morbird risk?
Low incidence disorders
Calculation for attack rate
Number of cases that develop/number exposed in a cohort (at risk)
i.e. incidence rate following acute exposure
Who created the filter model/pathways of care
Goldberg
What is the infant mortality rate?
Deaths under 1 year of age
What type of bias does blinding reduce?
Measurement
What is a factorial RCT?
When 2 or more interventions are compared separetly but also in combination and against a control
How are screening tools usually evaluated?
Cross-section studies
Describe Guttman scalogram
If you agree to one statement you agree to all statements beneath it. Also known as cumulative scaling.
What is Guttmann scalogram useful for?
Hierarchical constructs e.g. political views
Describe positive skew
mean>median>mode
unit of coefficient of variation
%
Define coefficient of variation
the sample standard deviation divided by the sample mean of the data set
Where are the mean and mode in skewed distributions?
Mode are always on the shorter side
Mean are always on the longer side
Calculation for probability of one of two events occurring which are mutually exclusive
the sum of their probabilities, minus the probability that both will occur
What is multistage sampling?
2x sampling events.
Initially, the larger primary units are randomly selected (e.g. municipalities in a country). Later individual units (households) within primary units are randomly selected for measurement
What is disproportionate stratified random sampling?
If a subgroup of interest is extremely small, we may also use different sampling fractions (f) within the different strata to randomly over-sample the small group.
Within each stratum, all individuals have an equal chance of being selected for the study.
What is the target population?
The group to which we wish to generalize the results of a study.
What type of error can be caused by repeating test on the same set of data?
Type 1
Value of no difference for NNT?
Infinity
What is a necessary condition for Validity?
Reliability
Values of Cronbach’s alpha
-infinity to +1
What does Cronbach’s alpha measure?
Internal consistency
When is multiple regression used?
when we have one dependent (Y) variable and many independent (X) variables.
Purpose of multiple regression?
to find an equation that best predicts the Y variable as a linear function of the X variables.
What type of test is multiple regression?
Multivariate
Requires parametric data
What is a prerequisite for regression?
Correlation
What value of kappa implies perfect disagreement?
-1
What is kappa a measurement of?
Agreement
Degrees of freedom for chi square?
(number of rows-1) X (number of columns-1)
What does multivariate mean?
Multiple independent variables
IV and DV in logistic regression?
DV: categorical
IV: categorical, continuous or discrete
What tests can be used for paired data?
Wilcoxon rank sum
Sign
Paired t test
What is Wilcoxin used for?
Non-normal data from large samples
What does degrees of freedom mean?
The extent to which a set of observations are ‘free’ to vary
What test is used to compare two survival curves?
Log rank test
What does Cochrans criteria refer to?
Criteria that must be fulfilled if the chi-square test is to be used for testing statistical significance
What are Cochrans criteria?
- All expected values in each cell have a frequency count = 1 i.e. non-zero values
2 At least 80% of total cells must have expected value of = 5
For which data type ca we use weighted kappa?
Ordinal
Assumptions for using ANOVA?
- Parametric distribution
- Equal variance amongst tested groups
- Independent observations
IV and DV for ANOVA
IV: categorical
DV: continuous
IV and DV for one-way ANOVA
IV: categorical
DV: continuous
What must be normally distributed for regression to be used?
DV
Calculation for Kappa
Kappa = Observed agreement beyond chance / Potential agreement beyond chance
What is the coefficient of determination (R^2)?
the proportion of the total variation in the dependent variable that is accounted for by the regression model (independent variable)
What does R^2 close to 1 imply?
that most of the variability in the dependent variable is explained by the regression model built using the predictors.
What is repeated measures ANOVA used for?
to compare paired observations within subjects across multiple groups.
What is the chi square equivalent for non-parametric data?
McNemars
What is the chi square equivalent for paired data?
McNemars
What is the type of sampling where every nth member is picked?
Systematic (non-random)
What is the name of the graph that is used to identify potential heterogeneity in which effect size is plotted against a measure of precision??
Galbraith plot
Level of non-significant difference for mean?
0
What is the normally accepted type 1 error rate?
5% (0.05 aka p value)
What is the square root of the variance?
SD
Which parametric test is used to compare 3 or more groups?
One-way ANOVA
IV and DV of logistic regression
DV: binary
IV: one or more
IV and DV for multiple linear regression?
1 DV
2 or more IV
What does multiple linear regression suggest?
Several IV together predict single DV
What is logistic regression used to predict?
Relationships between binary DV and several IV
When is logistic regression used?
If DV is binary
Calculation for population attributable risk
It is calculated by multiplying the attributable risk by the prevalence of exposure to the risk factor, and can be expressed by the formula PAR = Attributable risk X Proportion of population exposed.
What is attributable risk?
Attributable risk is the difference in risk between the exposed and the non-exposed.
Calculation for population attributable fraction?
PAF = Pe(RR - 1)/(1 + (Pe(RR - 1))), where Pe is the proportion of the population exposed to the factor and RR is the unadjusted relative risk of the outcome
What is the attributable risk?
Risk value attributable to exposure is the attributable risk.
How does one calculate the attributable risk?
. It is derived from obtaining the difference between control and experimental event rate.
Calculation for PAR?
It is derived from multiplying ARR from a study data with the proportion of a population that is exposed.
PAR = ARR X proportion of population exposed.
In which type of study is OR used?
Case control
What is incidence density sampling?
It involves matching each case to a set of individuals (usually >1) that are at risk of the exposure at the time of case occurrence.
What is needed in a study if the prevalence of a condition is low?
Larger sample size
What happens to sensitivity and specificity when two tests are used in sequence?
Loss in sensitivity
Gain in specificity
What happens to sensitivity and specificity if two tests are used in tandem?
Gain in sensitivity
Loss of specificity
What leads to high false positives?
Low-cut off
What happens if there is a high cut off for a disease?
Increased sensitivity
Decreased specificity
More false positives
What two things affect the predictive value?
Prevalence
If disease is rare - specificity