General Flashcards
What 3 components to a GLM
- systemic component - Y = X2
- random component - normal distribution, poissan, negative binomial, gamma
- link between the two- 1 (normal linear regression) or not 1, puts bend in the line
how do residuals appear in a standard linear regression model
vs
in a possion or negative binomial regression
Standard linear model - assume the reiduals are non normally distributed
in other, residuals form a poissan or negative binomial dsitribution
what is OLS?
what is the equation for this
Ordinary least squares (OLS) regression include things like:
- linear regression
predictor = c + Bx
what is screening?
establishes someones eligibility for the trial
includes
- informed consent
- eligibility assessment
demographics
medical history
what is a screening ID
soethign given to all n who are screened for the study, regardless of whether they are eligible for the trial or not
those not elkigible or not wishing to continue with the study are classified as “screen failures”
define adverse event
any unexpected medical occurence that may or may not be related to the product
define adverse reaction
unexpected response to an IMP that is related to the dose administered
there must be a reasonable possibility of the event being caused by the IMP i.e. the relationship cannot be ruled out
what is reference saftey information (RSI)
a list of medical terms detailing the ARs that are expected from an IMP. This information must be reviewed when assesseing a SAR for expectedness
Serious adverse event (SAE)
any unexpected response that
- results in death
- is life threatening
- requires inpatient hospitalisation or prolongation of existing hospitalisation
- results in significant disability/incapacity
- consists of congenital anomaly or birth defect
- anything else that jeopardises n or requires intervention to prevent one of the above consequences
serious adverse reaction (SAR)
an adverse event that has reasonable possibility to be caused by the IMP
Suspected unexpected Serious Adverse Reaction (SUSAR)
It is not expected (not consistent with reference saftey information (RSI)), but classed as serious and related to IMP
PICO?
▪ Population – Who should be included?
▪ Intervention - What is the intervention?
▪ Comparator - Comparing to what?
▪ Outcome - What is “good”?
how do we measure precision or undcertainty wuantitively?
o standard error
o confidence interval
o P value
what is standard error
a measure of the variability (SD) of sample means in repeated samples from the same population
quantifies how far from teh true (but unknown) population the sample estimate is
what are CI
o Plausible range of values for the true population value
o In fact, 95% of the times we compute a 95% CI, it will
contain the true (but unknown) population value
what kind of uncertainty to CI reflect
only uncertainty owing to random sampling
Non-sampling errors i.e. biases in design, conduct or analysis not
accounted for
relative risk or risk ratio (RR)
Is the ratio of the risk of a given event in one
group of subjects compared to another group
relative risk reduction
– (1-RR) x 100%
– The proportion of the initial or baseline risk
which was eliminated by a given
treatment/intervention or by avoidance of
exposure to a risk factor
odds ratio (OR)
Is the ratio of the odds of a given event in one
group of subjects compared to another group
absolute risk reduction (ARR)
The difference in risk of a given event, between
two groups
if a drug reduces the risk of a bad outcome from 50% to 40%, the ARR = 0.5 - 0.4 = 0.1
number needed to treat (NNT)
It is defined as the number needed to treat in
order to prevent one additional adverse event
(e.g. death)
the NNT = 1/ARR = 10.
hazard ratio
which is the study of the time it takes for an event of interest, such as death or disease progression, to occur.
compares the risk of the event occurring in two different groups, such as a treatment group and a control group, over a specified period of time.
Mediation analysis carried out using what technique?
Structual equation modelling.
are pilot studies and feasibility studies the same thing?
All pilot studies are feasibility studies but not all feasibility studies are pilot studies
who are the MHRA?
Medicines and Healthcare products Regulatory Agency
if treatment is an drug, then saftey moitoring is the jurisdiction of the UK regulator (MHRA)
What is the yellow card scheme
after a medical product has been lisenced for use in UK, this scheme is used for collecting and monitoring hte saftey of products
relies on voluntary reporting of side effects of drugs from healthcare providors or wider population
used to signal whether a drug might need to have its saftey checked again
which trials involving drug does the saftey reporting fall under the jurisdiction of the MHRA
RCT involving a drug
clinical trials ivolving IMP
Do we have to report all adverse events
yes, requirment within UK regulations to report unexpected AE’s or SAEs.
how are AE’s reported
MHRA provides specific monitoring and reporting requirments for clinical trials involving IMP
what is pharmacovigilance
fancy name for The continuous monitoring of the safe use of
medicinal products in clinical trials
do we need to include all AE’s in the expadited report?
No. depends on how much information is known about current treatment
if we already know a lot about drug. then shoudl focus on including important things, not collating a huge dataset of events that we already know about and expected.
when would we collect all vs not all AE’s
collects all if early phase drug trial. Less important when drug and AE’s associated are alrerady well documents. Maybe dont collec tthings that we expect.
how do we decides which AE’s to report and how is this decided
clinical decision but this needs to be approaved by the MHRA.
> How much is known about the intervention?
* Is it being used in a new population?
* Is it a drug being used at a different dose or in combination with other drugs?
> How much data already exists regarding the
safety of the intervention and/or the protocol
driven procedures?
can 2 protocols have the same saftey reporting requirments?
no
when AE is reported how do we determine causality ? if this was from IMP
- Determined with reference to the applicable product
information - Summary of Product Characteristics (SmPC) – licenced product
- Investigator Brochure (IB) – unlicenced product
- Protocol - if not IMP, but e.g., a behaivural intervention
difference between Serious Adverse
Reaction and Serious Related Event
Serious and related in a CTIMP is a Serious Adverse Reaction
Serious and related in a non CTIMP is a Serious Related Event
what do we refer to when dertermining whether AE was expected?
> For non drug trials this would be the protocol
For drug trials this would be the Reference
Safety Information (RSI)
* SmPC for licensed drugs (section 4.8)
* Investigator Brochure (IB) for unlicensed drugs
Serious Unexpected and Related Event
- classed as serious
- related
- Unexpected in a nonCTIMP
who do you report saftey issues to and within what time frame
The REC need to know about Serious, Related and Unexpected events (Non-CTIMP) within 15 days of sponsor receiving notification
The MHRA need to know about Suspected,
Unexpected, Serious Adverse Reactions (CTIMP)
* Fatal or life threatening – within 7 days
* Non-fatal and non-life threatening – within 15 days
what annual reports are needed?
> For non-CTIMPs generate an annual progress report (APR) and submit to REC
> For CTIMPs ensure a Development Safety Update Report (DSUR) is produced and submitted annually to the CA(s) & REC
