Epidemiology and biostatistics Flashcards
Why carry out randomisation?
-Avoids selection bias
-Successful randomisation makes different groups have similar characteristics at baseline
Why carry out blinding?
-To reduce or eliminate bias
-Bias may be intentional or unconscious
Types of blinding
What is unblinding?
-The disclosure to patients or study team of which treatment participants received during trial
What is intention to treat (ITT) analysis?
-ITT analysis includes every subject who is randomised according to randomised treatment assignment
-Ignores non-compliance, protocol deviations, withdrawal, and anything that happens after randomisation
-ITT analysis maintains prognostic balance generated by randomisation
-Estimate of treatment effect is generally conservative
-A better application of ITT approach is possible if complete outcome data are available for all randomised subjects
What is per protocol analysis
-A subset of ITT population who completed the study without any major protocol violations
What is skin prick testing?
-Small amount of allergen placed on skin and skin is pricked
-If child is allergic to any allergens, small bump (hive) will develop within 5-15 minutes
-This disappears after about 30 minutes
What is randomised control trial (RCT)?
Intervention study:
-Choice of treatment/intervention allocated randomly
-Typically randomised to new vs current or placebo treatment
-‘Gold standard’ in research studies
Why randomise in RCTs?
-Ensures patient’s characteristics don’t affect which treatment they receive
-Unbiased
-Treatment groups balanced
-Any differences in outcome can be attributed to treatment received
-Fair test of efficacy
What are cohort studies?
-Observational study
-Subjects observed in natural state (real world)
-Investigates causes or factors associated with disease (or condition)
-Selects group of individuals
-Follow up to monitor disease state and possible risk factors over time
-Usually prospective (but retrospective designs may be used)
Advantages of cohort studies
-Data is collected prospectively, so estimates are less likely to be biased than those from a case-control
-RCTs can only investigate potential benefits for ethical reasons
-Imbalance between exposed and unexposed can be corrected in the analysis
-Cohort studies can look at exposure to treatment and harm
Example of cohort study
Framingham study
-Population based, observational cohort
-Investigated the epidemiology and risk factors for cardiovascular disease
-Showed that high blood pressure and high blood cholesterol are major risk factors for CVD
-Helped develop QRISK
What are case control studies?
-Observational study - no intervention
-Subjects observed in natural state (real world)
-Investigate causes or factors associated with disease
-Selects group with disease: ‘cases’
-Choose comparator group without disease: ‘controls’
-Compare cases and controls with respect to possible risk factors - usually retrospective
Limitations of case control studies
-Choice of control group affects comparison
-Data reported by subjects or from records - usually retrospective so may be incomplete, inaccurate or biased
-But often quick to do and inexpensive
-Evidence from case control studies may be used in planning future research
What are cross sectional studies?
Observational study
-Subjects observed in natural state (real world)
-Collect data for each subject at one point in time
-Like a cohort study, but without follow up
What are cross sectional studies useful for?
-Measuring prevalence of a disease/condition
-Surveys/opinion polls: attitudes, views, behaviours
-Not useful for looking at cause and effect
Correlation vs causation
-Cross sectional studies show correlation, not causation
-Cohort studies can show causation
Ranking of study designs
1) RCT is gold standard (interventional)
2) Cohort study (observational)
3) Case controlled study (observational)
4) Cross sectional study (observational)
Why summarise data?
-To simplify data and be able to identify normal and abnormal values
-To monitor data quality
-To check for invalid or missing entries
-To describe characteristics of participants in a study
-Before doing a complex analysis
2 types of quantitative data
Continuous:
-Lies in a continuum, can take any value (age, weight)
Discrete:
-Can only take certain values, integers
Qualitative or categorical data
-Individuals fall into one of several categories
Binary: simple form of categorical data (e.g. male/female)
2 types of Qualitative (categorical) data
Ordinal data:
-Can be arranged in numerical order from smallest to largest
-Quantitative data always ordinal
-Some categorical data have inherent ordering so are ordinal (e.g. stage of disease)
Nominal:
-No ranking value (e.g. blood type, eye colour)
Ways of summarising continuous data
Centre of data
-Mean (arithmetic average)
-Median (middle value when data ranked)
Spread of data
-Range (min, max)
-Standard deviation (SD - shows dispersion of data)
-Variance SD^2
-Interquartile range
Which summary of data to find centre of distribution for quantitative data?
-Continuous data with symmetric distribution -> Arithmetic mean
-Continuous data with skewed distribution -> Median (if positive skew consider geometric mean)
-Discrete data -> Median unless range of data is large enough for calculation of a mean to be sensible
Which summary of data to find spread of distribution for quantitative data?
-Continuous data -> Use standard deviation (SD)
-Continuous data with skew - Consider using IQR as well
-Continuous data -> range (min to max) is useful in addition to SD if space allows
Graphical displays of continuous data
Histrogram
-Shows shape of distribution, range, middle
Box and whisker plot
-Upper quartile - top of box
-Lower quartile - bottom of box
Common distribution vs Positive skew vs Negative skew
Graphical displays of categorical data
-Bar charts
-Pie charts
