Basic Stats for the FRCS (Urol) Exam
-
Upload
drnikhilvasdev -
Category
Health & Medicine
-
view
1.457 -
download
5
description
Transcript of Basic Stats for the FRCS (Urol) Exam
Nikhil Vasdev, David Thomas Department of Urology Freeman Hospital Newcastle upon Tyne
Help in understanding clinical evidence that influences our day to day practice
Essential to have a thorough understanding to function as a successful urologist
Important to validate literature
Essential for the FRCS (Urol) exam
As Urologist we must be aware of a number of different ‘biases’ present in current literature which include
Media
Pharmaceutical Industry
Colleagues
Terminology 1. Prevalence – Total number of cases in a population at a given time
2. Incidence – The number of new cases in a population per unit time
3. Prevalence = Incidence X disease duration
4. Prevalence > Incidence = Applicable for chronic disease
5. Prevalence = Incidence – for acute disease (e.g. common cold)
Sensitivity
Number of true positives divided by number of all people with the disease
“Sensitivity = Positive in disease”
Specificity
Number of true negatives divided by number of all people without the disease
“Specificity = Negative in health”
Positive Predictive Value (PPV) Number of true positives divided by number of people who tested positive for a disease
The probability of having a condition, given a positive test
Negative Predictive Value (NPV) Number of true negatives divided by number of people who tested negative for the
disease
The probability of not having the condition given a negative test
Important points Unlike sensitivity and specificity, PPV is dependent on the prevalence of the disease
The higher the prevalence of a disease, the higher the positive predictive value of the test
Table 1 + -
+ A B
- C D
Disease Te
st
Sensitivity = A Specificity = D ______ ______ A + C B + D PPV = A NPV= D _______ _______ A + B C + D
Meta-analysis
Case-control study
Cohort study
Clinical trial
Meta-analysis Pooling of data from several studies (often via a literature search) to achieve a greater statical power
Main disadvantage – Cannot overcome limitations of individual studies or bias in study section
Case-control study Observational study (Retrospective)
Sample chosen on the basis of presence (cases) or absence (controls) of disease
Information collected about risk factors
Cohort study Observational study
Sample chosen on the basis of presence or absence of risk factors
Subjects are followed over time for development of disease
Clinical trial
Experimental study
Compares benefits of 2 or more treatments
Highest quality study = RANDOMIZED CONTROL TRIAL
Statistical technique for combining results of several studies into a single numerical estimate
Validity of MA depends on the quality of the systematic review on which it‘s based
Results are usually displayed with C.I., p values and a Forest plot ‘
A forest plot (or blobbogram) is a graphical display designed to illustrate the relative strength of treatment effects in multiple quantitative scientific studies addressing the same question. It was developed for use in medical research as a means of graphically representing a meta-analysis of the results of randomized controlled trials
A Bias is defined as when an outcome is more likely to occur than another
Selection Bias Subjects choose group
Recall Bias
Knowledge of presence of disorder alters recall by subjects
Sampling Bias
Subjects are not representative
Late look bias
Information gathered at an inappropriate time
Blind studies
Placebo responses
Crossover studies
Randomization
Phase 1: evaluates safety with increasing dose
Phase 2: early work on possible benefits/ efficacy
Phase 3: Formal evaluation (RCT)
Phase 4: Safety reporting in use
Table 1 + -
+ A B
- C D
Disease E
xpo
sure
RR = [ a / a+b] ________ [c / c + d]
“PROSCAR more than halves the risk of developing acute urinary retention and the need for surgery”’
Urologists had different points of view regarding: “the 48% to 57% relative risk reduction promoted and the 1.9% to 2.4%
absolute risk reductions actually observed in the median risk of AUR and surgery, respectively” [PLESS; MTOPS]
Table 1 + -
+ A B
- C D
Disease E
xpo
sure
Experimental event rate (EER) = A / A+B Control event rate (CER) = C /C+D Relative risk = EER /CER
Table 1 + -
+ 42 (2.8%)
1471
- 99 (6.6%)
1404
Retention F
inas
teri
de
RRR = Risk difference = 2.8% = 57% _____________ ____ Baseline difference 6.6% ARR = CER – EER = 6.6 – 2.9 = 3.8
Table 1 + -
+ 42 (2.8%)
1471
- 99 (6.6%)
1404
Retention F
inas
teri
de
NNT = 1 = 1 = 26 ____________________ ________________ Absolute Risk Reduction 0.038
Absolute risk of a disease is your risk of developing the disease over a time period. We all have absolute risks of developing various diseases such as heart disease, cancer, stroke, etc. The same absolute risk can be expressed in different ways. For example, say you have a 1 in 10 risk of developing a certain disease in your life. This can also be said to be a 10% risk, or a 0.1 risk - depending if you use percentages or decimals.
Relative risk is used to compare the risk in two different groups of people. For example, the groups could be smokers and non-smokers. All sorts of groups are compared to others in medical research to see if belonging to a group increases or decreases your risk of developing certain diseases. For example, research has shown that smokers have a higher risk of developing heart disease compared to (relative to) non-smokers.
Null (H0) Hypothesis of no difference
E.g. . There is no association between the disease and the risk factor in the population
Alternative (H1) Hypothesis that there is some difference
E.g.. There is some association between the disease and the risk factor in the population
Type 1 (α) Stating that there is an effect or difference when none exists (to mistakenly accept the
experimental hypothesis but reject the null hypothesis)
E.g. . You “saw” the difference that did not exist [Convict an innocent man]
P value of < 0.5
This indicates there is a less than a 5% chance that the data will show something that is
not really there
Type 2 (β) Stating that there is NOT an effect or difference when one exists (to fail to reject the
null hypothesis when in fact the null hypothesis is false)
E.g. . You “did not see” the difference that does exist [Setting a guilty man free]
Probability of rejecting the null hypothesis when it is in fact false
Power depends on
Total number of the end points experience by the population
Difference in compliance between treatment groups
The power of a test is the probability that a study of a given size would detect as statistically significant a real difference of a given magnitude
“If you increase the sample size, you increase the power. There is power in numbers”
In statistical significance testing, the p-value is the probability of obtaining a test statistic at least as extreme as the one that was actually observed, assuming that the null hypothesis is true
A measure of the effect of chance within a study
It is not the probability that the result of the study is true or correct
Normal = Gaussian distribution = Bell Shaped
Bimodal
Positive skew (Mean > Median > Mode)
Negative skew (Mean < Median < Mode)
It shows the trade-off between sensitivity and specificity (any increase in sensitivity will be accompanied by a decrease in specificity)
The closer the curve follows the left-hand border and then the top border of the ROC space, the more accurate the test
The closer the curve comes to the 45-degree diagonal of the ROC space, the less accurate the test
The area under the curve is a measure of test accuracy
The Kaplan–Meier estimator also known as the product limit estimator, is an estimator for estimating the survival function from life-time data
The term "survival" is a bit misleading; you can use survival curves to study times required to reach any well-defined endpoint (e.g., re-occlusion of a grafted blood vessel, first metastasis, discharge from the hospital).