Clinical sciences Flashcards
what is the p value and whats its sig value
Stat sig p value = 0.05 = probability that the null hypothesis is true/that you can reject the null hypothesis
if a trial estimate is 1.4 what does that mean
Estimate 1.4 = 40% more likely to be in remission (or 1.4x)
if a CI is described as (95% between 0.8 and 2) what does this mean
CI = estimate of precision (95% between 0.8 and 2) = 20% worsening of outcome and 100% improvement
when do you use
- chi squared
- t test
- mann whitney u test
Chi squared: categorical data
T test: continuous variable that’s normally distributed (and in 2 groups)
Mann Whitney U test: continuous variable that’s NOT normally distributed (and is in 2 groups)
case control study
- explain
- when do you use
- what bias?
for rare diseases. Start with the outcome. but get recall bias!
cohort study
- what for
what is it
what bias
Cohort: for common diseases. Start with the exposure/risk. Get channeling bias (if haven’t randomised ppl to a tx, there’s often a reason they ended up with that one)
type 1 error
- what is it
what causes it
Type 1: when we get a false positive
= incorrectly concluding that there’s a stat sig difference in a dataset
— (causes: failure to adjust for confounders, inadequate blinding, not performing an intention to treat analysis, poor randomisation)
type 2 error
what is it and what causes it
Type 2: when we get a false negative
= incorrectly conclusing that there’s no stat sig difference in a dataset
— (sample size is too small; classically inadequately powered study)
Aka boy who cried wolf: 1st time he cried wolf, village came running and he had made it up = false positive
but 2nd time, there is a wolf but village doesn’t believe him = false negative
sensitivity equation
= the number of pts with a positive test who have the disease divided by all pts with the disease
if high wont miss many ppl with the disease
= few false negatives aka good for ruling out diseases (like d-dimer in pe)
true positive = new test +ve & ref test +ve
Sensitive so it picks up lots of trash as +ve too
- False neg = new test -ve and ref test +ve
specificity equation and what is it
Specificity = true neg / (true neg + false positive)
= the number of pts who have a negative test and dont have the disease dividedby the number of patients who dont have the disease
= few false positives; so if its positive you probably have the disease = relatively diagnostic
- the test is specific to the disease, only positive when you have it
- true neg = both neg
= False +ve = new test +ve and ref test -ve
NNT equation & what is it
NNT = 1/absolute risk difference
- ARR = the event rate in the control arm - event rate in treatment arm
- aka 10% of control group relapsed vs 5% of treatment arm
- Absolute risk diff = 10-5 = 5% = 0.05
- so NNT = 1/0.05 = 20
endothelin what is used in primary pulmonary hypertension
endothelin antagonist
p53 is a what
tumour suppressor
positive and negative predictive values
Negative predictive value = chance of testing negative and not having the disease = true negs/all the people who tested negative
PPV: true positives/all who tested positive
complement deficiencies predispose to ? - c1 inh protein def? - C1q, C1rs, C2, C4 deficiency C3 deficiency C5 deficiency: C5-9 deficiency:
- C5-9 def predisposes to n.meningitidis infections
- C1q, C1rs, C2 and C4: immune complex disease (eg SLE, HSP)
NOTES
C1 inhibitor protein deficiency: hereditary angioedema
C1q, C1rs, C2, C4 deficiency (classical pathway components)
C3 deficiency: recurrent bacterial infections (esp encapsulated)
C5 deficiency: Leiner disease (recurrent diarrhoea, wasting and seborrhoeic dermatitis)
C5-9 deficiency:encodes the membrane attack complex (MAC)
relative risk equation
A total of 100 patients received the new drug whilst 100 patients received a placebo. Of the patients taking the experimental drug, 60 felt improved symptoms compared to 20 patients in the control group.
whats the RR of the drug improving migraine syms
Relative risk = EER/CER = ratio of a risk happening in the experimental group vs control group
EG 60/100 in expt group had improved symptoms, 20/100 in placebo group did - so 60/20 = 3
boy, brother & their male maternal cousin what is it most likely?
mitochondrial
HLA A3 B51 B27 (3) DQ2 DQ8 DR2 (2) DR3 (3) DR4 (2)
HLA-A3: haemochromatosis
HLA-B51: Behcet’s disease
HLA-B27
• ankylosing spondylitis
• reactive arthritis
• acute anterior uveitis
HLA-DQ2/DQ8: coeliac disease
HLA-DR2
• narcolepsy
• Goodpasture’s
HLA-DR3
• dermatitis herpetiformis
• Sjogren’s syndrome
• primary biliary cirrhosis
HLA-DR4
• type 1 diabetes mellitus
rheumatoid arthritis - esp DRB1 gene (DRB104:01 and DRB104:04 hence the association with DR4)
expressivity v penetrance
· Expressivity = extent to which a genotype shows its phenotypic expression in an individual
· Penetrance = proportion of a population of individuals who carry a disease-causing allele who express the related disease phenotype.
turner’s
- denoted as?
- features
- most common cardiac abn
· 45X0
· Short, webbed neck, shield chest, primary amenorrhoea, high arched palate, horseshoe kidney, high gonadotrophins
· Most common cardiac defect: bicuspid aortic valve (also ass w aortic root dilatation & coarctation of aorta)
NOTES
Features
• short stature
• shield chest, widely spaced nipples
• webbed neck
• bicuspid aortic valve (15%), coarctation of the aorta (5-10%)
• primary amenorrhoea
• cystic hygroma (often diagnosed prenatally)
• high-arched palate
• short fourth metacarpal
• multiple pigmented naevi
• lymphoedema in neonates (especially feet)
• gonadotrophin levels will be elevated
• Hypothyroidism
• horseshoe kidney (most common renal abn)
• increased incidence of autoimmune disease (especially autoimmune thyroiditis) and Crohn’s disease
fragile x
- inheritance
- type of disease
fts in males and females
Features in males • learning difficulties • large low set ears, long thin face, high arched palate • Macro-orchidism • hypotonia • autism is more common • mitral valve prolapse
Females (1 fragile chromosome and 1 normal): normal to mild
trinucleotide repeat
standard error of the mean equation
SEM = standard deviation / square root (number of patients)
So the SEM gets smaller as the sample size increases
breast feeding CI
- drugs
- other 2
Drugs to avoid: • abx: ciprofloxacin, tetracycline, chloramphenicol, sulphonamides • psych: lithium, benzodiazepines, clozapine • aspirin • carbimazole • methotrexate • sulfonylureas • cytotoxic drugs • Amiodarone
Other CI
• Galactosaemia (babies can’t metabolise galactose - so can’t BF as it has lots of lactose in - they need non-lactose formula)
• viral infection
X-linked recessive diseases (14)
• Androgen insensitivity syndrome • Becker muscular dystrophy • Colour blindness • Duchenne muscular dystrophy • Fabry's disease • G6PD deficiency • Haemophilia A,B • Hunter's disease • Lesch-Nyhan syndrome • Nephrogenic diabetes insipidus • Ocular albinism • Retinitis pigmentosa • Wiskott-Aldrich syndrome Chronic granulomatous disease: X-linked recessive in >70% (but some other patterns of inheritance)
key feature of a normal distribution (and any others)
KEY BITS
· Mean = mode = median
· standard deviation (SD) is a measure of how much dispersion exists from the mean
NOTES
Properties of the Normal distribution
• symmetrical i.e. Mean = mode = median
• 68.3% of values lie within 1 SD of the mean
• 95.4% of values lie within 2 SD of the mean
• 99.7% of values lie within 3 SD of the mean
• this is often reversed, so that within 1.96 SD of the mean lie 95% of the sample values
the range of the mean - (1.96 *SD) to the mean + (1.96 * SD) is called the 95% confidence interval
- Ie If a repeat sample of 100 observations are taken from the same group 95 of them would be expected to lie in that range
Standard deviation
- the standard deviation (SD) is a measure of how much dispersion exists from the mean - SD = square root (variance)
equations
- LV ejection fraction
- stroke volume
- CO
- systemic vascular resistance
- pulse pressure (2 factors that increase it)
· LV ejection fraction = (stroke volume / end diastolic LV volume) x100%
NOTES
Stroke volume = end diastolic LV volume - end systolic LV volume
Cardiac output = stroke volume x heart rate
Systemic vascular resistance = mean arterial pressure / cardiac output
Pulse pressure = Systolic Pressure - Diastolic Pressure
- Factors which increase pulse pressure
○ a less compliant aorta (this tends to occur with advancing age)
increased stroke volume
