lecture 1 - introduction to evidence based medicine Flashcards
this is a statistical method that shows the relationship between two or more variables
regression analysis
There is no current evidence that _____ have a CVD benefit
DPP4 inhibitory
_____ and ____ have shown CVD benefit (modest and higher in higher risk groups)
SGLT2i and GLP1a
this is the conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients
evidence based medicine
describe the hierarchy of evidence
1 = top of hierarchy (best)
- meta-analysis of randomized control trials (RCT’S)
- individual RCT
- observational studies (patient important outcomes)
- basic research (test-tube, animal/human physiology)
- clinical experience (non-systematic clinical observation)
what are the three factors that affect evidence based medicine (EBM)?
- best evidence
- patient values
- clinical expertise
this is one of the three factors that affect EBM ad consists of religious & moral beliefs along with preferences & rights
patient values
this is one of the three factors that affect EBM and consists of clinical trials and systematic reviews
best evidence
this is one of the three factors that affect EBM and consists of professional judgement and experience
clinical expertise
this is just a number, but can help to describe data and help us make decisions. can be descriptive: numerical information about an object or event derived from a sample (study or trial) from a population; can facilitate inferences about a population when only part of those data (sample) are actually observed
statistic
what are the three common scales of measurement for variables in medicine
- nominal
- ordinal
- numeric
this is the simplest scale of measurement for variables in medicine; data fits into categories with no particular order; there is no actual measurement - just a count; often dichotomous or binary (yes/no; disease/no disease); generally described in percentages or proportions.
nominal
this scale of measurement for variables are also called qualitative observations or categorial observations
nominal
this scale of measurement for variables has an inherent order to the categories; summary statistic is the median. is often used in assessment of patient risk; different between 2 adjacent categories is not the same throughout the scale (e.g. going from stage 1 to 2 of cancer may not be as severe as going from stage 3 to 4)
ordinal
this scale of measurement for variables is also called quantitative observations
numerical
this is one of the types of numerical scales; has a value on a continuum (e.g. age, weight, blood pressure)
continuous scale
this is one of the types of numerical scales; values are integers (e.g. # of fractures, # of medications)
discrete scale
what are the two summary statistics?
mean and standard deviation
this is arithmetic average. used with numerical values and not normally with ordinal values
mean
what’s the formula for mean?
the sum of x / n
where x is the individual observations and n is the number of observations
this is the middle observation when the observations are listed from smallest to largest
median
when the number of observations is odd, the median is the middle number, how do you find the median for an even number of observations?
the median is the average of the values on either side of the middle
this is the value that occurs most frequently
mode
this kind of graph occurs when outlying values are small; mean < median
left/negatively skewed
this kind of graph occurs when outlying values are large; mean > median
right/positively skewed
what occurs when the mean and median are similar with regards to a graph?
symmetric distribution
in what type of graph should you use mean?
symmetric
in what type of graph should you use median?
ordinal or numerical data that is skewed
what are the different measures in spread?
- range
- standard deviation/variance
- coefficient of variation
- percentiles
- interquartile range
this is a measure of spread; it is the difference between the smallest and largest observation. minimum and maximum may also be given
range
this is the measure of the variation of our data from the mean
standard deviation
______ is the statistic before the square root is taken for the standard deviation
variance
this is the measure of relative spread; this allows comparison of relative variation of distributions measured on different scales
coefficient of variation
what is the formula for the coefficient of variation?
CoV = standard deviation (s)/mean (x-bar) x 100
this is the percentage of a distribution that is equal to or below a particular number; median is 50th percentile; physical growth charts for children is a common usage
percentile
this is defined as the difference between the 25th and 75th percentile; this describes the middle 50% of the distribution regardless of the shape
interquartile range
this is used with mean with symmetric data
standard deviation
this is used with median for ordinal data or skewed numerical data
percentiles and interquartile range
tabular presentations consist of nominal and ordinal data presented as ________ or ____________
proportions or percentages
this type of table facilitates simultaneous examination of multiple distributions e.g. explore bivariate association between gender and surgery
contingency tables
what are different ways to organize/visualize numerical data
- stem and leaf plots
- five number summary
- box plots
- grouped frequency tables
what does the stem consist of in a stem and leaf plot?
all but the right most digit (e.g. 483, the stem is 48)
what does the leaf consist of in a stem and leaf plot?
the right most digit (e.g. 483, the leaf is 3)
true or false: the stem of a stem and leaf plot should be written from smallest to largest
true
what is the purpose of the stem and leaf plot? (what is it helping us visualize?)
the collection of leaves will have the general shape of the distribution r
this helps to show the location and spread of the data; it displays a full range of data (min and max), displays a common range (25th percentile/Q1 and 75th percentile, Q3) and displays a typical value
five number summary
in a box plot or box and whisker plot, what are the upper and lower hinges made with?
1st and 3rd quartile
in a box plot or box and whisker plot, the _____ is the line that is found within the box
median
in a box plot or box and whisker plot, symmetry is evaluated by the symmetry of the hinges with respect to the median; if the hinges are equidistant from the median the data is ____________
symmetrical
in a box plot or box and whisker plot, symmetry is evaluated by the symmetry of the hinges with respect to the median; if the upper hinge is further away from the median, the data is ________ skewed
positively
in a box plot or box and whisker plot, symmetry is evaluated by the symmetry of the hinges with respect to the median; if the lower hinge is further away from the median the data is ______ skewed
negatively
in a box plot or box and whisker plot the spread of data may also be shown by _____; these are drawn from the upper and lower hinges TO the largest/smallest non-outlying values
whiskers
in a modified box plot, what is the boundary for outliers?
1.5 x interquartile range from the box
group observations on variable are placed into contiguous, non-overlapping _______; this is a term used in statistics when we are given a continuous series.
class interval
*class means a group of numbers in which items are placed such as 0-10, 10-20, 20-30.
*class interval refers to the numerical width of any class in a particular distribution
this indicates how often a specific kind of event occurs within the total number of observations
relative frequency
used to determine the number of observations that lie above (or below) a particular value in a data set; the collection of all previous frequencies together
cumulative frequency
how can you construct grouped frequency tables?
- group observations on variable into a continguous, non-overlapping class interval (bins)
- place each observation into only one bin
- tabulate frequency of observations in each bin - can calculate relative frequency (proportion/percentage)
- can also tabulate cumulative frequency
what is the general rule for how many class intervals you should have?
5-20
grouped frequency distributions are displayed visually as a _________. unlike bar graphs, these are generally joined as they represent a continuous distribution (not frequency of categories)
histogram
the area of the bar of a histogram is proportional to the _________
frequency
Relative frequencies can also be grouped as a histogram, looks the same, but has a different ___ scale (AUC = 1 or 100%)
Y
this is created by linking the mid-points of successive bins.
frequency polygon
this is a form of frequency distribution that represents the sum of a class and all of the classes below it
cumulative frequency distribution
when this is on the y-axis, it allows us to estimate the median based on looking at the graph
cumulative frequency distribution
what is the effect of having larger samples?
larger samples = more observations = more bins = smaller bins
