Stats Flashcards
In a randomised controlled trial of a new influenza vaccine, 100 patients were
randomly allocated to receive the vaccine and 100 to receive placebo. During
follow up, 20 patients who had received the vaccine contracted influenza,
compared with 30 patients who had received placebo.
Which effect did the vaccine have on the relative risk of contracting influenza?
RR = (20/100)/(30/100) = 0.67, which implies a 33% reduction in risk
what type of plot shows publication bias in a metaanalysis?
how do you interpret a funnel plot?
Funnel plot
symmetrical = no bias
assymettrical = bias
prospective cohort studies are at the highest level of data. only above which is RCTs and and systematic reviews
1 group of data, trying to find variation, which test do you use?
wilcox ranked signed test!
testing the same group twice and normally distributed = paired t test
two different groups and normally distibuted = unpaired T test
man whitney U test -> two independent populations- breastfeeding vs bottle-feeding mothers). The data are also not normally distributed
ANOVA = mean of 3 sets of data
NNT (number needed to treat formula)
NNT = 1/absolute risk reduction
absolute risk reduction = control event rate - experimental event rate
phase 0 = behaviour in human body
phase 1 = safety assessment
phase 2 = efficacy
phase 3 = comparison to other drugs/rct
phase 4 = post marketing surveillance
standard deviations are
68-95-99.7 (99.7 being standard devations below mean)
The average weight loss of a patient following a new type of bariatric surgery is 18 kg. The standard deviation of weight loss is 3kg. Assuming the weight loss is normally distributed, what percentage of patients will loss between 9 and 27 kg?
99.7
A graphical display of continuous data where the values have been categorised into a number of categories is a ?
A plot of the estimate of a patient’s survival showing decreasing survival with time?
Histogram
Kaplan-Meier survival plot
in review, it’s found the trial only enrolled patients with high socioeconomic status who have regular access to healthcare.
Which type of bias most likely affected the outcome of this clinical trial?
selection bias!!
Performance bias is incorrect. Performance bias occurs when there is a systematic difference in the care provided to the groups being compared, other than the intervention itself.
Expectation bias (Pygmalion effect) - observers may subconsciously measure or report data in a way that favours the expected study outcome.
A small study looks at the weight of patients diagnosed with type 2 diabetes mellitus. Overall 64 patients were reviewed. The average weight was 81 kg, with a standard deviation of 12 kg. What is the standard error of the mean?
1.5
Standard error of the mean = standard deviation / square root (number of patients)
type 1 vs type 2 error
A type I error occurs when the null hypothesis is rejected when it is true. The opposite has occurred in this case.
A type III error is not widely used. This is where the null hypothesis is rejected correctly but for the wrong reason. = there is acc a difference
ach patient receives one test, producing 1,700 positive results. 680 of those with a positive test are later confirmed to have the disease.
Calculate the positive predictive value (PPV) of the test. = 40%
TP / (TP + FN ) = sensitivity
Specificity TN / (TN + FP)
Positive predictive value TP / (TP + FP)
Negative predictive value TN / (TN + FN)
You are asked to design a study to assess whether living near electricity pylons is a risk factor for childhood leukaemia. What is the most appropriate type of study design?
Case-control study - compares a group with a disease to a group without, looking at past exposure to a possible causal agent for the condition
Cohort study is incorrect because, whilst it could theoretically be used, it would be impractical and inefficient for studying childhood leukaemia. A cohort study would require following a very large number of children for many years to observe enough cases of leukaemia, given its relative rarity. This would be time-consuming and expensive.
