Flashcards in Test 2 Deck (30)
What are the components of quality health care?
3: patient centeredness
Define evidence based practice:
The explicit, conscientious and judicious use of current best evidence in making decisions about care of individual patients(and populations)
Aim of EBP?
Integrating of best
- research evidence
- clinical expertise
- patient values
What is the EBP process?
2: Ask an answerable clinical question
3: Aquire the evidence
4: Appraise the evidence
5: Apply the evidence
What are the possible research designs when there is a control and intervention group?
Either Randomised control trial OR
Controlled Clinical Trial
What research design is...When 2 large groups, treated and nontreated are followed for years or even decades to count how many develop a particular disease
Which research design is used when the HISTORIES of a group of people with a condition is compared with a group of people without the condition - in order to find suspected casual factors:
Research design of, interviewing women asking if they have had a baby in the past few months and if they took a specific drug during pregnancy?
Cross sectional studies
What kind of data is mild, moderate, severe?
What kind of data is gender, eye colour, burning-tingling-shooting?
What kind of data is height, weight, temp?
What kind of data is #students, counts?
Stating the null and alternative?
________(Intervention) had not significant effects on _______(what is being measured) of the _______(describe participants)
When testing hypothesis with p-values:
1. State significance level (usually alpha = 0.05)
2. state p values for what effects are being measured
3. Is the p value greater or smaller than significance level?
- if P < 0.05 = reject the null
- if P > 0.05 = fail to reject the null
What does a p value of 0.05 mean?
When assuming the null (with significance level 0.05)
You'd obtain the observed difference or more in 5% of studies due to random sampling error
Why would someone set the significance level at 0.01?
When we want to make sure chance plays as small a role as possible
Usually done with a large sample and when there is high risk
Eg. Deadly medications
Probability diagnostic test is positive in patients who do have the disease
Eg. Sensitivity of 0.85 = there is a 85% chance that if the test is positive that you do have the disease
Draw out diagnostic accuracy table
Probability diagnostic test is negative in patients that do not have the disease
Eg. Specificity of 0.75? There is a 75% chance that if the test is negative the individual does not have the condition
What is the false negative rate?
Probability the test will fail to detect disease in a patient with disease
What is false positive?
Probability test is positive for someone without the disease
Use the info given to display the results of the diagnostic test in a table (2x2 table):
Y axis = test result
X axis = disease state
Both positive then negative
Only fill table with counts at this stage
Eg. 12 TP, 14 FP u knooo
What is the formula for sensitivity?
TP/(TP + FN) X 100
What is the formula for specificity?
(TN/TN + FP) X 100
How do you recommend things based on sensitivity and specificity?
Say which one is low, and which one is slightly better
Say how close to 100 eg. Too far from, fairly close
Therefore I would say this is a) not a good test b) a good test for ___
In stats what does the 68-95-99.7 rule mean?
Just draw a normal distribution with 1-3 standard deviations mannnnnn
"When data is normally distributed the 68-95-99.7 rule refers to approximately how many of your scores will fall between a certain number of standard deviations
1. It is a probability value that tells you how much chance played a role in that research
2. The p-value is used to reject or fail to reject the null(the null is the starting point for hypothesis testing)
3. The p-value is related to the sample in study(draw circle diagram with population and sample and p value and confidence intervals)
Explain confidence intervals:
1. Confidence intervals are estimated range of values within which population parameter is going to lie
2. Confidence level comes from alpha Eg. Significance level of alpha = 0.05 = 95%, 95% confident true pop mean lies between stated values of lower and upper confidence limit
3. The narrower CI more precise
How do you know a study is a randomised control?
If it states random allocation anyyywhereere aye