Stats and such Flashcards
exposure odds ratio
diseased/healthy exposed divided by diseased/healthy unexposed
Interpretation of an odds ratio
under the rare disease assumption (100%), we can substitute odds for risk, and odds ration for relative risk
What is the relationship between prevalence and incidence
Prevalence=Incidence x disease duration
What is the target population
Population to which inferences from the study are to be made.
Source population
Population from which study subjects (cases and non-cases) are selected
Effect modification
Effect modification is when the effect of A on B depends on C
e.g. the effect of smoking on laryngeal cancer is greater in men than in women EM
as opposed to, the effect of coffee on laryngeal cancer is confounded by smoking
Measures of disease frequency
Measures of disease frequency describe the absolute risk of a disease.
Prevalence,
Incidence rate/Incidence density, and
cumulative incidence/incidence proportion
measure of association
Risk ratio (relative risk , incidence rate ratio), A tool to compare disease rates in two populations
Odds ratios
Risk difference (attributable risk)
The population attributable risk (rate) (PAR) is the risk in the total study population that is attributable to the presence of the exposure
Logistic regression
Logistic regression can describe the relationship between a categorical outcome (response variable) and a set of covariates (predictor variables). The categorical outcome may be binary (e.g., presence or absence of disease) or ordinal (e.g., normal, mild and severe). The predictor variable(s) may be continuous or categorical.
Type 1 error
incorrect rejection of a true null hypothesis. (False positive result)
Type 2 error
is the failure to reject a false null hypothesis (incorrect acception of a false null hypothesis) (false negative result)
4 Measures of variability
There are four different measures of variability, the range, interquartile range, variance, and standard deviation.
Variance
The average squared difference of the deviance from the mean 1,2,3,4,5 medel 3 -2 -> 4 -1 -> 1 0 -> 0 1 -> 1 2 -> 4 = 10 10/n-1 = 10/4= 2,5
Standard deviation
square root of the variance
Sensitivity, specificity, PPV, NPV
Sensitivity=sant testpositiva/alla sant positiva
specificity=sant testnegativa/alla sant negativa
PPV= sant Testpositiva/ alla testpositiva
NPV= Sant Testnegativa/ alla testnegativa
Whats the difference between paulines study and yours
Basically the same incidence for TNFi (66 vs 62)
I have an active comparator
Dramtic increase in melanoma age-standardized incidence (doubling since 2000, 5 times higher than 1975), almost no change in metastatic melanom and death from melanoma
In the european collaborative project, the inidence rate in Swedish TNFi treated was double that of UK-treated (the two largest registers)
No increase of in situ melanoma in paulines study
Active comparator important in your studies (stuyd 3)
Results were very stable comparing the different bionaive comparators
Potentiell bias i both TMX/AI
Samtidigt kan det ju finnas orsaker kopplade till ledvärk som gör att man 1) byter preparat, 2) alternativt att om man inte har utvecklat jobbig ledvärk av ett av preparaten så är man kanske mer villig att testa det andra också.
Adjustments in the cc study
Betr brca som riskfaktor för senare RA är det eg inte lika viktigt att justera, vår fråga är ju här om OR för brca före RA liknar HR för RA efter brca, oavsett om det förstnämnda är confoundat eller ej.
Genetic overlap
Both RA and Breast cancer are gentically complex diseases. GWAS studies have indeed identified many candidate genes, but the effect sizes are generally quite small
R2
proportion of the variance in the dependent variable that is predictable from the independent variable(s
AIC
AKaika information criterion, a measure of the fit of the model, comparing two models, lower value is better
Hazard Hazard ratio
Instantaneous risk, ratio of the hazard rates
Properties of the normal distribution
Normal distributions are symmetric around their mean.
The mean, median, and mode of a normal distribution are equal.
68% of the area of a normal distribution is within one standard deviation of the mean.
Approximately 95% of the area of a normal distribution is within two standard deviations of the mean.
Channeling between bDMARDs
We see that mortality is higher in non-TNFi cohorts, especially among those that start a non-TNFi as their first bDMARD.
We also see that there is channeling of patients concerning the risk of malignancy (most of which is handled by age/sex + disease severity), but we see that when we compare tnfi switchers to second tnfi or non-tnfi, there is very little channeling beyond age/sex
Compliance vid TMX-behandling
I en registerstudie var 30% ickekompliant
Biverknignar AI TMX
TMX, svettningar trombos
AI, ledvärk, osteoporos, torrhet i slidan,
