RESS Flashcards
Define hypothesis
Testable statement that describes an observation
Define theory
well confirmed explanation of nature but remains falsifiable
Define Health Outcome
Impact healthcare activities have on people
Focusses on what results from what is done
Name the 3 health outcomes
Record Based
Biological and Clinical Outcomes
Clinician and Patient-reported outomes
Give an example of record based health outcome
Mortality, incidence
Give an example of Biological and clinical outcomes
Lab results, BMI, Blood pressure
Give an example of Clinician and patient-reported outcomes
Symptom score
HRQOL
Define Incidence
New cases of that disease arising in a population over a defined period of time
Number of new cases/Number at risk (in a given period of time)
Define Prevalence
Cases of people living with the disease
Number of people with a disease/Number of people in the population (at a given time)
Define Case Fatality Rate
Number of people who die from the disease/Number of people with disease in period
Define Mortality Rate
Number of people who die from the disease/Number of people who die in period of time
Define Stratum
Subgroup of your individuals in the sample
Define Risk
Number of new cases/Number at risk
Define Odds
Probability of event/Probability that event doesn’t occur
Define and explain relative risk
Risk in exposed group/Risk in unexposed group
=1 means no benefit or harm
>1 means harmful effect
What is standard error used to calculate?
The confidence intervals
Define P-value
probability that such a test specific could take a value at least that extreme
Pearson Correlation Coefficient
Measured between two numeric variables
When should PCC not be used?
Non Linear relationship
Outliers
Distinct Sub Groups
When is Spearman Rank Used?
When data is not normally distributed, when a variable is ordinal or a small sample size
Define Linear Regression
Fitting a straight line to points on scatterplot
Define Censoring
Leaving the study before the event occurs or entering after
Define failure
The time the person leaves the study is known, may be due to event or people lost
Define Right Censored
People did not reach failure before end of study
Define Left Censored
Not certain what happened to people before they entered study
Give a common health related status
Death
Give three keyword search methods
Truncation (eg. anorexi = anorexia, anorexic) Adjacency Searching (relate two things together) Wildcards (eg. behavio?ral = one character or none) (eg. behavio#ral = one character)
Give the common Boolean Searches
AND - requires both words to be returned
OR - either terms or both will be returned
NOT - first term is searched and removes any with second term
How would you summarise the data of normally distributed data?
Mean and Standard Deviation
How would you summarise the data of non normally distributed data?
Median and IQR
What graph would you use to describe the distribution of a continuous variable?
Histogram
Define Sensitivity
Percentage of people who are correctly identified in having the disease
TP/TP+FN
Define Specificity
Percentage of people who are correctly identified in not having the disease
TN/TN+FP
Define Positive Predictive Value
Probability of someone having the disease given a positive test result
TP/TP+FP
Define Negative Predictive Value
Probability of someone not having the disease give a negative test result
TN/TN+FN
Define Null Hypothesis
Assumes no effect in the population
Define Standard Deviation
Measure of spread of a distribution
Define Statistic
Sample estimate from population
Define Standard Error
Measure of the precision of the sample mean
Define Validity
Measure accurately what it is meant to measure
Eg. BMI rather than weight as measuring obesity
Define Reliability
Give same result when retested
Define Responsiveness
Can detect real changes when they occur
Eg. Continuous QoL scale rather than categorical
Define Continuous Data and an example
Numerical value with no limitation on values
Eg. Weight, regardless of whole units or not
Define Discrete Data and an example
Numerical Value with limitations on values
Eg. number of hospital admissions
Define Ordinal data and an example
Ordered category
Eg. Likert’s scale, stage of disease
Define Nominal data and an example
Non ordered category
Eg. Blood group, sex
