Midterm Flashcards
Efficacy
The extent to which a specific intervention, procedure, regimen, or service produces a beneficial result under ideal conditions
Effectivenes
A measure of the extent to which a specific intervention, procedure, regimen, or service when deployed in the field in routine circumstances, does what it is intended to do for a specified population
Where does the evidence come from?
- Research
- Textbooks
- Internet
- CE Courses
- Experts
Hierarchy of Evidence Quality
- Systematic reviews and meta-analysis
- Clinical trial (in humans)
- Longitudinal cohort study
- Case control study
- Descriptive and cross-sectional study
- Case report and Case series
- Personal opinion, subjective impressions, anecdotal accounts
ADA definition of evidence-based dentistry integrates 3 things
- Assessment of clinically relevant scientific evidence
- Clinical expertise
- Needs and preferences of the patient
Sponsoring agency for journals (4)
- Learned society
- Professional organization
- Scientific publisher
- Commercial publisher
Impact Factor =
# citations to articles appearing in journal ------------------------------------------------------------ # articles published
3 kinds of papers published in primary journals
- Research reports
- Reviews of the literature
- Commentaries
Research report allows for 3 things
- Assess observations
- Repeat experiments
- Evaluate intellectual processes
The content of an article should be:
New, true, important, and comprehensible
Components of research report:
- Title
- Authors
- Date of submission/acceptance
- Abstract
- Intro
- Materials and Methods
- Results
- Discussion
- Conclusion
Clinical trial in humans allows you to see
Cause and Effect
Cohort Studies
Group is defined based on their exposure to a suspected risk factor for disease (no cause/effect)
Case-Control Studies
Group is defined based on whether they do or do not have a disease
Cross-sectional Surveys
“Snapshot” in time. Disease and exposure are assessed together.
Can tell prevalence, not incidence
Case Reports and Case Series
Describe a single patient
Variable
An element, feature, or factor that is liable to vary or change
Mathematics
A quantity that during a calculation is assumed to vary or be capable of varying in value
Computing
A data item that may take on more than one value during the runtime of a program
Nominal Data
The classification of an observation according to the group to which it belongs. Ex: gender, political party, marital status, state of residency
Ordinal Data
A measurement scale based on the classification of an observation according to its relationship or other observations. Ex: poor-fair-good scale
Continuous Data
Real Numbers. Interval (temperature) or ratio (mass/time)
Mode
- What is it
- What data is best with it
- Most frequent measurement
2. Nominal data
Median
- What is it
- What data is best with it
- Middle
2. Ordinal data
Mean
- What is it
- What data is best with it
- Affected by?
- Average
- Continuous
- Outliers
IQR
How spread out the data is
Variance
Average value of distance from the mean of all of the variables. Square of SD
SD
Positive square root of the variance
Coefficient of variation
Measures the percentage of the spread
100 x SD/mean
Standard Error of the Mean
Several samples
Calculate mean
Calculate SD of the means
SD is used to measure
Variability of individual subjects/entities
SE is used to assess
How accurately a sample mean reflects a population
Bar charts are good for
Nominal and ordinal (categorical)
Histograms are good for
Continuous data
Box and Whisker
- Box represents
- Line in the box represents
- Whiskers extend to
- Circles represent
- IQR
- Median
- The remaining data
- Outliers
Dot plots
Continuous data in groups
Scatter plots
Two continuous variables
Probability
The relative likelihood of an event occurring, measured on a scale from 0-1. Sometimes discussed in %
Normal Probability Distribution
A density function assuming a bell shaped curve
For a small sample use:
T-distribution
For a bigger sample use:
Z-score
Central limit theorem and normal probability function allow us to
Infer about the population
Z-score percents of population
- 3% in 1 SD
- 5% in 2 SD
- 7% in 3 SD
Research hypothesis
A prediction based on the theory being tested. Language based statement that we are trying to prove
Null hypothesis
There is no difference. We are trying to disprove this with research
Alternate hypothesis
All possibilities that the null doesn’t cover. There is a difference.
Dependent Variable
What we measure
Independent Variable
What we think affects the outcome
Type I error
Rejecting the null when it is true
Type I error is also called
alpha or p-value
Type II error
Accepting the null when it is false
Type II error is also called
beta
Alpha and beta are
Inversely proportional
If we increase sample size, what happens to alpha and beta?
They go down
Power
The chance that we will find significance
Power and sample size are
Directly related
Methods to increase power (6)
- Increase the type I error
- Increase the sample size
- Increase the deviation from the null hypothesis you are willing to tolerate
- Decrease variability
- Use a directional alternate hypothesis
- Use the most efficient statistical test
Statistical Decision Making
- Research question
- Define variables
- State null and alternate
- Choose statistical test
- Determine type I error that you will tolerate
- Conduct the experiment
- Calculate the experiment
- Calculate the test statistic
- Determine the type 1 error (p value)
- Conclude
