Terminology Flashcards
Descriptive studies
Record activities, observations or events
Rare to provide info to evaluate new treatments
Eg case reports, case studies, population studies/cross-sectional studies
Case series
Tracks patients with a known exposure or similar treatments, or examines their medical records for exposure and outcome
Cross-sectional studies
Information about a population at a slice in time
Explanatory studies
Seek to find info about the causes of illness or the effectiveness of treatments
Observational studies
Studies of the course of disease and health events, then the results are used to reach conclusions
Types:
Identify individuals known to have a disease or problem, then looking for factors in the past that might have led to the problem
Dealing with individuals who don’t have a problem, but following them over time to study characteristics that might be related to development of a disease or problem.
Case control study
Individuals who ar known to have a disease or problem of interest and compares to others who aren’t affected
Cohort study
Selecting a starting point and collecting information as you go
Essentially a follow up study
Look at the causes of disease - identifying individuals with some sort of exposure and following them over time to see what happens to them
Provides info about the course and prognosis of diseases
Nature of group influences generalizability
Interventional studies
Specifically examine the effect that some intervention might have on patients
Eg social support, New med, new procedure to treat illness
Evidence level rankings
1 or A:RCT
II-I: CT but not random
II-2 or B: cohort. Or case-control studies
II-3. Or C: case series
III or D: expert opinion or descriptive studies
Nominal data
Categorical. Data
Given names
No level of severity, just different grouping. Eg blood type
Ordinal data
Categorical
Describe a progression. Of something
Observation put in order
Eg cancer stage
Binary data
Yes/no
Dead//alive
Continuous variables
Any value in a range
Height, weight, etc
Discrete data
Can only take certain values
Doses given,, completed weeks of gestation
Qualitative
Descriptions
Describe qualities of the. Information
Quantitative.
Can. Be measured with a numerical value.
Mode
Most common value within a group
Standard deviation
Spread of data
Mean of the mean
Tells how tightly all the various examples are clustered around the mean in a set of data
Confidence interval
Upper and lower values that will include 95% of data from a sample
Probability distribution
Way of showing the chances that any value will be found in a population
Eg normal distribution
Null hypothesis
No difference between two interventions
Alternative hypothesis
Real difference in the effectiveness of interventions
Statistical significance
%5% chance (1:20) that the difference found between two interventions in the study are due to random chance
95% certainty that the difference is real
P-value
Probability that the alternative hypothesis is correct
Chi squared probability test
The sum of the squared difference between observed and expected data, divided by the expected data in all possible categories
Chi Squared = (o-e)^2/e
Used with independent groups
Paired t-test
Used to determine whether the mean difference between two sets of observations is zero OR
Used to compare 2 population means where you have two samples in which observations in one sample can be paired with observations in the other sample
Used to determine if there is a sig diff between mean in two groups
Need the mean difference, standard deviation of each group, and the number of data values of each group
Used in case-control studies or repeats-measures designs
Eg before and after observations on the same subjects
A comparison of 2 different methods of measurement or two different treatments where the measurement/treatments are applied to the same subjects
Absolute risk
Probability that a certain event will occur in a specified population
Absolute risk difference/risk reduction
Difference in risk for an event or outcome between an exposed population and an unexposed population
Relative risk
Used in RCT and cohort studies
Ratio of two risks (or probabilities), usually the risk of an event or disease in an exposed group and the risk in an unexposed group
Odds ratio
Ratio of two mutually-exclusive events
Eg have pre-eclampsia, do not have pre-eclampsia
Type 1 error
Rejecting the null hypothesis when it is true
Type II error
Not rejecting a null hypothesis when the alternative hypothesis is the true state of nature
-usually when sample size too small
Unpaired t test
Compares two independent samples drawn from the same population
Mann-Whitney U test
F test
One way analysis of variance using total sum of squares
Compare three or more sets of observations on a single sample
Pearson’s r
Product moment correlation coefficient
Assesses the strength of the the straight line association between two continuous variables
Multiple regression by least squares method
Describes the numerical relation between a dependent variable and several predictor variables (covariates)
Incidence
Number of new cases of a condition or disease in a given population over a given time
