Block 34 H&S Flashcards
Who manages the blood transfusion service?
- the blood transfusion service is managed and co-ordinated by NHS blood and transplant - national organisation resp for blood donation, collection, testing, processing and dist
What is the NHSBTS ?
- NHSBT is a special health authority in England that oversees the provision of blood and organ donation and transplantation services.
- It operates under the oversight of the Department of Health and Social Care (DHSC) and reports to the NHS England and NHS Improvement
Blood donation centers?
- NHSBT operates a network of fixed and mobile blood donation centers across England.
- These centers provide facilities for voluntary blood donors to donate blood, plasma, and platelets.
Labs in the NHSBT?
- Donated blood undergoes testing for infectious diseases, blood group typing, and compatibility testing at specialized laboratories operated by NHSBT.
- Processing facilities prepare blood components such as red blood cells, platelets, and plasma for distribution to hospitals.
hospital blood banks?
- NHSBT supplies blood and blood products to hospitals across England through its regional distribution network.
- Hospital blood banks receive and store blood products from NHSBT and provide them to patients as needed for transfusion.
- They also perform compatibility testing for patients and monitor transfusion reactions.
NHSBT specialised services?
- NHSBT provides specialized services for patients with specific transfusion needs, such as neonatal and pediatric transfusions, immunoglobulin therapy, and rare blood product provision.
using blood products safely?
- blood compatibility - ABO and Rh typing as well as crossmatching
- following protocols
- coumentation
- education - risks and signs of complications
- adversr reaction management
- post-transfusion monitoring
Aim of the NHS diabetes prrevention programme 2016?
Aim of the programme is to reduce the incidence of type 2 diabetes and its associated complications
NHS DPP identifies?
It identifies patients high risk of developing type 2 diabetes and enrolling them into a programme over a 9 months period
Three
NHS DPP is focused on?
The programme focused on achieving weight loss, increasing physical activity and improving diet
Where is the nHS DPP available?
only England
Who is eligible for the NHS DPP?
Individual above 18 of age and with HbA1c in the non-diabetic hyperglycaemia(42-47mmol/mol) is eligible for this programme
confidence interval =
range of values that most likely contains the true value of the population
which 3 factors determine the width of a confidence interval?
- variation within the population
- sample size
- confidence level
Mean
a sample with little variation means a sample taken
the sample taken will have a mean similar to that of the population
more variation ->
wider CI
Larger sample size ->
a larger sample size has more variation so it decreases the CI
A 95% confidence interval means that…
- in 95 out of 100 samples the value will fall between the upper and lower values specified by the CI
CI graph
95% is the general consensus
risk of disease =
number of people who have the disease/ all exposed
risk of disease in non exposed =
number of people who developed the disease in the unexposed group/ total unexposed
Attributable risk =
- added risk of developing an outcome based on exposure
- difference in risk between the exposed group and unexposed gr
AR equation
attributable fraction =
- percent of an outcome that could possibly be prevented if a risk factor was to be removed
- a.k.a as attributable risk percentage
if the attributable fraction for smoking and lung cancer is 70% this means that
70% of lung cancers are related to smoking
equation for AF?
risk in exposed - risk in unexposed
/
risk in exposed
alt equation for AF that can be helpful for observational studies =
(RR-1)/ RR x 100
normal dist?
- bell shaped curve
- is a probability distribution
- continoud and symmetric around the central value - mean
Normal dist - mean, median, mode?
all placed at the center
variance =
- shape of the curve
- refers to spread of distribution
- often expressed as the SD
Equation
SD =
sq route of variance
what follows normal dist?
- BP
- birth weight of babies
Increasing mean vs increasing variance?
- increasing variance makes the graph flatter and wider
- changing the mean moves the curve to the left or to the right but doesn’t change the shape
% of values that fall within 1 SD of the mean?
68%
probability of values that fall within 2 SDs of the mean?
95%
% that fall within 3 SD of the mean?
99
Sensitivity =
- how good a test is at finding disease
- proportion of affected indiv correctly identified by the test
Sensitivity equation =
true positives/ all affected
tests with high sensitivity help us rule ? disease
OUT
SnOUT
Highly sensitive tests are useful for…
screening tests
specificity related to how well the test…
identifies disease free individuals
specificity =
prop of unaffected indiv corretcly identified by the test
Sp equation =
true negatives/ all affected
all affected means the true negatives and the false positives
risk ratio is the
- measure used to compare the risk of an event occuring between 2 groups - often an exposed and unexposed grs
risk ratio is also known as the
relative risk
relative risk is used to assess the
strength of association between an exposure and outcome
RR is usually used t interpret results of which study type?
cohort studies
RR equation
- number of people with disease in the exposed group/ all exposed
- number of people in the unexposed gr w disease/ all unexposed
divide these 2 values
RR of 1 means that
- there is no association between disease and exposure
RR > 1 means that
there is an increased risk of disease occuring with exposure
Risk ration <1 means that
there is reduced chance of disease occuring w exposure
Which is preferred between RR and OR?
RR preferred as its based on probability
for rare diseases where prev is low, the odds ratio can
approximate the risk ratio
RR = 5 means that
there is a 5x inc risj
RRR=
proportion of risk reduced when comparing the exposed and the unexposed group
equation for the RRR?
1 - risk ratio
Absolute risk reduction =
difference in risk between the unexposed and the exposed group
Equation for ARR?
- Like relative risk but you minus instead of dividing the risk in each group
- so the risk in the unexposed - the exposed group
odds ratio is the
- odds of exposure in people who have the condition vs the odds of exposure in the controls
Odds interpretation?
- same as RR but also look at the confidence interval
- CI should be above 1 for statistical significance
odds ratio equation =
exposed free of disease/ those without disease and without exposure
A/ C divided by C/ D
When are risk and odds ratios used?
- RR: cohort
- OR: case control studies
Rare disease rule?
risk ratio can be used in place of odds ratio fir case control studies
What needs to occur for OR to be considered stat sig?
CI needs to be above 1
What is a positive likelihood ratio?
- tells us how much more likely a positive result is in those with the condition
positive likelhihood ratio is the odds of a disease being present when
an indiv tests positive
PLR formula?
sensitivity/ 1 - specificity
Negative LR =
- can tell us how much more likely a negative test result is in those without the condition compared to those w
NLR is the odds of…
disease not being present when an indiv tests negative
we want a low NLR for a good test
NLR formula?
1 - sensitivity/ specificity
What does a LR of 1 indicate
- test is not useful
- result is unable to distinguish between those w disease and those without
LR test of <1 indicates that
- the test for a disease is useful in ruling out disease as a negative result is more likely to occur in those without the disease
- negative LR -> ruling out
LR >10 indicates that
- the test is useful in ruling in disease as a positive result is more likely to occur in those w the disease
- positive LR -> ruling in
A LR of between 1 and 10 is
not useful. Can’t rule in or out disease
what do predictive values look at?
- ratio of patients correctly diagnosed to the apparent test results
PPV is the ratio that
- someone who has tested positive actually has the disease
PPV equation =
number of true positives/ all positives
x100 to get the %
NPV is the
chance of not having the disease when the test is negative
NPV equation
number of true negatives/ all negatives
prev =
total number of indiv in a population who have a disease at a specific period of time, given as a percentage
as prevalence rises,
- PPV rises
- NPV rises
as bc the prev is going up, more ppl have the disease and so are more likely to test positive
RCT design?
- ppt randomly allocated to intervention or control gr
- practical or ethical problems may limit use
cohort studie mechanism?
- observational and prospective
- ppts selected according to exposure and then followed up to see how many develop the disease or outcome
usual outcome measure in cohort studies?
relative risk
CCS are
observational and retrospective
what happens in CCS?
- patients with a condition - cases are identified
- they are matched with controls
- data is collected on past exposure looking for a cause of the condition
CCS usually use the
odds ratio
benefits of CCS?
- inexpensive, produce quick results
- useful for studying rare conditions
- of CCs?
prone to confounding
Cross sectional studies provide a
snapshot - sometimes called prevalence studies
CSS -
provide weak evidence for cause and effecr
bias definition =
- where one outcome is systematically favoured
selection bias?
- error in assigning people to groups which leads to differences between groups
recall bias =
- differences in recall
- particularly in case control studies
Example of recall bias?
E.g. a patient with lung cancer may search their memories more thoroughly for a history of asbestos exposure than someone in the control group
Publication bias =
- failure to publish results from valid studies as they showed a negative of uninteresting results
- e.g. meta analyses where negative results may be excluded
expectation bias =
- Only a problem in non-blinded trials.
- Observers may subconsciously measure or report data in a way that favours the expected study outcome.
Hawthorne effect?
Describes a group changing it’s behaviour due to the knowledge that it is being studied
lead time bias =
Occurs when two tests for a disease are compared, the new test diagnoses the disease earlier, but there is no effect on the outcome of the disease
length time bias?
- overestimation of survival due to excess of cases detected that are slowly proressing
lead vs length time bias?
- Lead-time bias is due to early detection. Remember the “d” in lead is for early detection.
- Length-time bias is due to slow cases being detected more often simply because they are slowly progressing. Remember the “g” in length is for slowly progressing.
validity =
accuracy of results
Reliability =
Reliability is used in statistics to imply consistency of a measure.
case control vs cohort in terms of retro and prospective?
cohort: either pro or retro
CCS: retro
What is the most appropriate design for diagnosis?
cross sectional = observational study that analyses data from a
population at a specific period of time
studies looking for aetiology?
- cohort
- CCS
cohort study =
ongitudinal study that follow a population, often one that has a
particular exposure – e.g. smoking
case control study =
= population study in which two existing groups differing in
outcomes are identified & compared based on basis of some supposed causal
attribution
prognosis studies ?
cohort
which study to use for treatment?
RCT/systmatic reviews
Evaluation studies?
SR/ MAs
Advantages of a cohort study?
- Best information about causation
- Able to examine a range of outcomes
- Good for rare exposure
Disadvantages of a cohort study?
- Long follow up = expensive & time-consuming
- Bad for long latency periods
- Can have different follow-up for exposed / non-exposed
- Confounders not recognized
advantages of CCS?
- Simple / easy to conduct = don’t require long follow ups
- Best for rare outcomes
- Good for long latent periods
Disadvantages of CCS?
- bad for rare exposure
- inferior to cohort
- recall bias
- confounding
selection bias =
- Error in assigning individuals to groups, leading to differences in
group’s qualities that may influence the outcomes - To make inferences from results, we require sample to be
representative of population = requires random sampling (or if
smaller group some stratified randomization)
Reasons for selection bias?
- Sampling -> selected subjects not rep of population
- volunteer - volunteers not rep of population
- non responder - responders not rep of population
randomisation?
- ensures unpredictable assigment to groups
- Purpose is to ensure that any confounding
characteristics are equally distributed between
the two study groups, avoiding selection bias
simple randomisation?
computer random number generator (may have chance bias)
stratified randomisation =
to make sure you have equal numbers of e.g. different severities of illness in each patient group
when something is precise the CI is?
narrow
internal validity =
How well study was conducted, taking
confounders into account & removing bias
external validity =
Generalizability -> how well the study can be applied to diff scenarios/ patients/ environment
what is face validity?
general impression of a test. A test has face validity if it appears to test what it is mean to.
content validity =
- refers to extent to which a test or measure assesses the full content of a subect area
criterion validity =
- comparisons of tests
- uses correlation coefficient
construct validity =
extent to which a test measures the construct it aims to
what does blinding prevent?
- prevents observer bias
- confirmatio bias
- expectation bias
observer bias =
form of reactivity in which researcher’s cognitive bias causes them to subconsciously influence the participants of experiment = could influence
extra quality of care to these patients
expectation bias =
= Observers may subconsciously measure or report data in a way that favours the expected study outcome
managing attrition bias?
- ITTA = All patient analyzed even if didn’t complete study = if dropped out either use data before drop out or include this in analysis as a limitation of the study / potential
Bradford Hill criteria to assess for causation?
- Strength = effect size
- Reproducibility = in concordance with other findings
- Specificity = other likely explanation for results
- Temporality = effect occurs after cause
- Biological gradient = greater exposure = greater effect
- Plausibility = logical mechanism by which effect is achieved
what is the gold standard for cause and effect?
- RCT
- but it’s not ethical to allocate patients to certain exposures that may be harmful
questionnaires are subjec to?
recall bias & response bias (a group with a certain condition may be less likely to respond)
what is confounding =
“Distortion (or potential for distortion) of association between outcome and exposure, by a third factor which has an association with both exposure &
outcome”
How is confounding controlled for?
if confounding has not been taken into account in a study?
if not, association between exposure and outcome could be biased = can stratify, or (more often) do Multi-Variable Regression
inappropriate measurement can lead to
type 1 or type 2 error
incomplete follow up can reduce?
power of the study
insufficient length of follow up ->
underestimation of outcome and exposure association
systematic reviews +
- speed
- ethics - can u justify the ethics or a new trial if trials are already available
- increased sample size -> improved power and precision
cross sectional studies?
- observational
- Source population is examined to see what proportion has the outcome of interest, or has been exposed to the risk factor of interest, or both.
- Looks at individuals within a population at a single point in time
cross sectional studies are most useful to investigate?
diagnsosi
ecological studies?
An observational study
- Data analysed at population/group level rather than individual
- Often used to measure prevalence/incidence of disease particularly when disease is rare
evidence pyramid
Number needed to treat?
- “Number of patients who need to be treated to prevent one additional adverse outcome in a given time”
Equation for NTT?
- 1 - ARR
- round up to full number
equation for NHH?
- “Number of patients who need to be treated to prevent one additional adverse outcome in a given time”
- 1+ ARR
- Round down to full number
P value
- Likelihood that the observed result is due to chance
- less than 5, statis sig
power of study =
- Power of a study is the probability of correctly rejecting the null hypothesis (power is increased by increasing sample size
Type 1 error =
- False positive - null hypothesis is incorrectly rejected
- not affected by sample size
Type 2 error =
- Null hypothesis is incorrectly accepted = false negative
- Decreased by smaller sample size
Phase 1 of a clinical trial?
- small gr
- assesses safety. pharmacokinetics and pharmacodynamics, side effects before moving on to larger studies
phase 2 studies?
- larger gr to assess dosing and efficiacy