Chapter 14 - Biostatistics Flashcards
Editor’s role
The editor of the journal selects potential publications and sends them to experts in the topic area for peer review.
Peer review’s role:
Assess
- the research design and methods
- the value of the results and conclusions to the field of study
- how well the manuscript is written
- whether it is appropriate for the readership of the journal.
Reviewers role:
- The reviewers make a recommendation to the editor to either accept the article (usually with revisions) or reject it.
- Data that contradicts a previous recommendation, or presents new information, can change treatment guidelines.
What are the 6 STEPS TO JOURNAL PUBLICATION
1) Begin with a research question:
Write a null hypothesis to answer the research question, such as: New drug is not as effective as current drug.
2) Design the study:
Is it randomized, placebo-controlled, a case-control or other type of study?
3) Enroll the subjects:
Assign to a treatment group or control group, or identify subjects belonging to a cohort or other group.
4) Collect the data:
- Prospectively (going into the future for a set period of time) or
- Retrospectively (looking back in time using medical records).
5) Analyze the data: Enter the data into statistical software; assess the results (e.g., risk reductions, confidence intervals).
6) Publish
Types of study data:
1) Continuous Data:
- Interval Data: No meaningful 0 (0 Celsius)
- Ratio Data: Meaningful 0 (HR: 0 bpm)
2) Discrete (Categorical) Data
- Nominal (Arbitrary order): M/F or Y/N
- Ordinal (Logical order): Pain scale (4 is not double 2)
Quantity
Quality
Measures of central tendency: (Define)
■ Mean:
- Average value
- Add up the values & divid the sum by the # of values.
- Preferred for CONTINUOUS data that is NORMALLY distributed
■ Median:
- The value in the middle when the values are arranged from lowest to highest.
- When there are two center values, take the average of the two center values.
- Preferred for ORDINAL data or CONTINUOUS data that is SKEWED (not normally distributed).
■ Mode:
- Value that occurs most frequently.
- Preferred for NOMINAL data.
SPREAD (VARIABILITY) OF DATA
■ Range:
The difference between the highest and lowest values.
■ Standard deviation (SD):
Indicates how spread out the data is, and to what degree the data is dispersed away from the mean
A large number of data values close to the mean has a smaller SD.
Data that is highly dispersed has a larger SD.
GAUSSIAN OR NORMAL DISTRIBUTION
- Characteristics
- Bell-shaped distribution
- Large sample sets of CONTINUOUS data
- Curve is symmetrical
- Most of values closer to the middle
- Mean, Median, Mode are at the CENTER and SAME value
- 68% of the values fall within 1 SD of the mean
- 95% of the values fall within 2 SDs of the mean.
SKEWED DISTRIBUTIONS
- Curve is not symmetrical
- 68% of data do not fall under 1 SD from the mean
- Mean, median and mode are not the same value
- Small sample size &/or there’s outliers in data
- The MEDIAN is a better measure of central tendency
- As the number of values inc, the effect of outliers of the mean dec
Data is skewed towards the outliers
1) Positive (Right) skewed:
- Data is more to the LEFT
2) Negative (Left) skewed:
- Data is more to the RIGHT
What are the types of endpoints to the study’s variables
- Clinical endpoint: hospitalization, death…
- Intermediate (Surrogate) endpoint: measuring CrCl to assess renal fcn
What are the Dependent & Independent Variables?
- Dependent: The outcome (hosp, death…)
- Independent: Chosen by researcher (to determine whether it has an effect on dependent variable or outcome), characteristics of pts, inclusion/ exclusion criteria (age, gender…)
Null hypothesis Vs Alternative Hypothesis
Ho: null
- no statistical difference
- should be rejected
Ha: alternative
- statistical diff
- Should be accepted
They are complementary
What is the alpha level or standard for significance?
- Max permissible error margin
- Set by researchers
- Commonly 5% or 0.05 (should be <0.05)
- The threshold of rejecting the null hypothesis
- If they want to chose a lower alpha (1%); requires:
- More data –> more expensive
- Larger ttmt effect
- It correlates to the values in the tails when the data has a normal distribution
What is the P-value and how is it used in assessing statistical significance?
1) Researcher determine alpha
2) Statistical tests are done to compare data
3) P-value is calculated; i need p-value < alpha
- if alpha is 5%,
p-value <5 –> reject null hypothesis and stat sig
p-value >= 5 –> accept null hypothesis and no stat sig
What is the Confidence Interval?
It shows 2 main things, what are they?
- Shows significance and precision
- CI = 1 - Alpha
If a=0.05, and p-value turned out to be < 0.05, then
I am 95% confident that the conclusion is correct and there is 5% chance that it is not –> STAT SIG
The result is stat sig if the CI does not cross 0 (diff) or 1 (ratio)
- Narrow CI –> high precision (preferable)
- Specific pt types will cause wider distribution in data
(Pts with high TG levels will experience a higher drop of TG)
Type I Error (False Positive)
- The alternative hypothesis was accepted and the null hypothesis was rejected IN ERROR.
- The probability, or risk, of making a type I error is determined by ALPHA and it relates to the CONFIDENCE INTERVAL.
CI = 1 - Alpha
- When alpha is 0.05 and a study result is reported with P-value < 0.05, it is statistically significant and the probability of a type I error (making the wrong conclusion) is < 5%
- You are 95% confident (0.95 = 1- 0.05) that your result is correct and not due to chance.
Type II Error (False Negative)
- Denoted as beta
- Null hypothesis is accepted when it should have been rejected.
- Beta is set by the investigators during the design of a study.
- Set at 0.1 or 0.2, meaning the risk of a type II error is 10% or 20%.
- The risk of a type II error increases if the sample size is too small.
- To decrease this risk, a power analysis is performed to determine the sample size needed to detect a true difference between groups.
Study Power
- Power is the probability that a test will reject the null hypothesis correctly (i.e., the power to avoid a type II error).
Power = 1 - beta
- As the power increases, the chance of a type II error (Beta) decreases.
- Power is determined by the number of outcome values collected, the difference in outcome rates between the groups, and the significance (alpha) level.
- If beta is set at 0.2, the study has 80% power (there is a 20% chance of missing a true difference and making a type II error).
- If beta is set at 0.1, the study has 90% power. A larger sample size is needed to increase study power and decrease the risk of a type II error.
Risk
- Probability of an event to occur when the intervention is given
Relative risk or risk ratio
Risk = (Number of subjects in group with an unfavorable event) / (Total number of subjects in group)
RR = (Risk in treatment group)/ (Risk in control group) x100
- RR= 1 (or 100%) implies no difference in risk of the outcome between the groups.
- RR> 1 (or 100%)implies greater risk of the outcome in the treatment group.
- RR< 1 (or 100%) implies lower risk (reduced risk) of the outcome in the treatment group.
ex: RR of HF progression was 57%.
Patients treated with metoprolol were 57% as likely to have progression of disease as placebo-treated patients.
A placebo-controlled study was performed to evaluate whether metoprolol reduces disease progression in patients with heart failure (HF).
A total of 10,111patients were enrolled and followed for 12months.
What is the relative risk of HF progression in the metoprolol-treated group versus the placebo group?
Calculate the risk of HF progression in each group. Then calculate RR.
METOPROLOL
- N = 5,123
- HF progression: 823
CONTROL
- N = 4,988
- HF progression: 1,397
Metoprolol Risk:
823/ 5,123 = 0.16
Control Risk:
1,397/ 4,988 = 0.28
RR = R/R = 0.16/0.28 = 0.57
57%
The RR of HF progression was 57%.
Patients treated with metoprolol were 57% as likely to have progression of disease as placebo-treated patients.
Relative risk ratio
- Indicates how much the risk is reduced in the treatment group compared to the control group.
RRR = (% risk in control group - % risk in treatment group) / % risk in the control group
OR
RRR = 1 - RR (use RR in decimal)
ex: The RRR is 43%.
Metoprolol-treated patients were 43% less likely to have HF progression than placebo-treated patients.
RR + RRR =
1
RR as likely
RRR less likely
The RR and RRR provide relative (proportional) differences in risk between the treatment group and the control group; they have no meaning in terms of absolute risk.
Absolute Risk Reduction
- Absolute risk reduction is more useful because it includes the reduction in risk and the incidence rate of the outcome.
- If the risk of nausea is reduced, but the risk was small to begin with (perhaps the drug caused very little nausea), the large risk reduction has little practical benefit.
- It is best if a study reports both ARR and RRR
ARR = (% risk in control grp) - (% risk in treatment grp)
interpretation:
The ARR is 12%, meaning 12 out of every 100 patients benefit from the treatment.
Said another way, for every 100 patients treated with metoprolol, 12 fewer patients will have HF progression.