Biostatistics & Pharmacoeconomics Flashcards
What’s the median/ 50th percentile?
Value in the middle of a ranked list
When do u use the median value?
When data is skewed
What’s the mode?
Value that occurs most in a data set
What’s a “normal distribution” also known as?
Bell shaped or Gaussian curve
Characteristics of Gaussian or normal distribution?
Mean, mode and median would all have the same (similar) value.
Curve is symmetric around the mean
Skew = 0 (zero)
What’s a negatively skewed distribution (asymmetric)?
The mean, mode and median are not the same (similar)
The mean is towards the left of the distribution (and not in the center)
What’s a positively skewed distribution (asymmetric)?
The mean, mode and median are not the same (similar)
The mean is towards the right of the distribution (and not in the center)
Whats the p-value?
Likelihood (or probability) that result obtained was due to chance
What does p-value < 0.05 mean?
There’s less than a 5% probability that the result was obtained by chance
When do we fail to accept (reject) the null hypothesis?
When the p-value is smaller than the predetermined significant level (say < 0.05), indicating that the observed result is highly unlikely under the null hypothesis
When’s Confidence Interval (CI) considered significant in odds ratio, risk ratio or hazard ratio that compares 2 grps?
When the CI doesn’t include 1
When’s Confidence Interval (CI) considered significant between grps (or within the same grp over time)?
When the CI doesn’t include 0
Is the measure of “statistical significance” the same as “clinical significance”
No
What’s a type I error?
Also known as “false positive”
When the null hypothesis is true, yet it is rejected in error
What’s statistical power?
1 - B (type II error)
What does a higher statistical power imply?
Means that we can be more certain that the null hypothesis was correctly rejected
What’s relative risk/ risk ratio / RR?
Probability of an unfavorable event occurring in the tx grp versus the control grp
Howz RR expressed?
Decimal or percentage
Formular of RR?
Risk = number of subjects with unfavorable event
Divided by
Number of subjects in that arm
RR = Risk in tx group
divided by
Risk in control grp
How do u interpret RR = 1?
RR = 1: no difference in risk btw the 2 grps
How do u interpret RR < 1?
The event is LESS likely to occur in the tx grp than in control grp
How do u interpret RR > 1?
The event is MORE likely to occur in the tx grp than in control grp
A study compared metoprolol v. placebo in heart failure pts over 12 months. If HF progression occurred in 28% of placebo-treated pts and in 16% of metoprolol-treated pts.
Calculate the RR
Interprete the RR
RR = tx Grp/ control grp
= 0.16/0.28 = 0.57 or 57%
T4, subjects treated with metoprolol were only 57% as likely as placebo-treated pts to have HF progression
What’s another formula for calculating RR?
RR = a/(a+b)
Divided by
c/(c+d)
In a group of 100 smokers, 40 developed lung cancer while 60 didn’t. In a similar grp of 100 non-smokers, lung cancer developed in 10 people.
Calculate the RR
Interpret the RR
RR = a/(a+b)
divided by
c/(c+d)
RR = 40/(40+60)
divided by. = 4
10/(10+90)
Means that smokers are 4 times more likely to develop lung cancer than non-smokers
Formula for relative risk ratio (RRR)
RRR = (% risk in control grp - % risk in tx grp)
divided by
% risk in the control grp
OR
RRR = 1 - RR
A study compared metoprolol v. placebo in heart failure pts over 12 months. If HF progression occurred in 28% of placebo-treated pts and in 16% of metoprolol-treated pts.
Calculate the RRR
Interprete the RRR
RRR = (28% - 16%)/ 28% = 0.43 OR (1 - .57)
There’s a 43% relative risk reduction in HF progression in pts being treated with metoprolol
What’s absolute risk reduction (ARR)?
Difference btw control grp event and tx grp events
ARR = (% risk in control grp) - (% risk in tx grp)
A study compared metoprolol v. placebo in heart failure pts over 12 months. If HF progression occurred in 28% of placebo-treated pts and in 16% of metoprolol-treated pts.
Calculate the ARR
Interprete the ARR
ARR = (% risk in control grp) - (% risk in tx grp)
ARR = 28% - 16% = 12%
This is the difference in risk that can be attributed to the intervention (drug)
What’s number needed to treat (NNT)?
NNT = 1/
(% risk in control grp) - (% risk in tx grp)
OR
NNT =. 1/
ARR (expressed as a decimal)
A study compared metoprolol v. placebo in heart failure pts over 12 months. If HF progression occurred in 28% of placebo-treated pts and in 16% of metoprolol-treated pts.
Calculate the NNT
Interprete the NNT
NNT = 1/
(% risk in control grp) - (% risk in tx grp). OR
NNT = 1/
ARR (expressed as a decimal)
NNT = 1/(0.28 - 0.16) = 8.3 round up always! = 9
T4 for every 9 pts who received metoprolol for 1 year, you would expect HF progression to be prevented in 1 pt
What’s odd ratio? (OR)
OR is the ratio of the odds of an event offering in the tx grp to the odds of an event occurring in the control grp
Measure of association btw exposure and an outcome
In a group of 100 smokers, 40 developed lung cancer while 60 didn’t. In a similar grp of 100 non-smokers, lung cancer developed in 10 people.
Calculate the OR
Interpret the OR
Odds of a smoker developing lung cancer = 40/60 = 0.67
Odds of a non-smoker developing lung cancer = 10/90 = 0.11
T4 OR = 0.67/0.11 = 6
Means that smokers are 6 times more likely to develop lung cancer compared to non-smokers
What’s the mean?
Average value of a sample distribution
What’s a positive correlation graph?
Downward sloping line from right to left
What’s a negative correlation graph?
Downward sloping line from left to right
What’s a dependent variable?
Outcome of interest, which may change in response to some intervention
What’s an independent variable?
The intervention
List types of discrete data
Nominal - consists of categories, where order is arbitrary e.g. Gender
Ordinal - consist of ranked categories, but differences btw categories are not equal e.g. Likert scales, NYHA functional class
List types of continuous data (infinite number of possible values, such as height, weight)
Interval data - used to measure continuous data with legitimate mathematical values. The difference btw 2 consecutive values is consistent along any pt of the scale, but the zero pt is ARBITARY e.g, Celsius temp scale
Ratio - has equal intervals btw values and a MEANINGFUL zero pt (height, weight, time, length)
List types is study types
Observational study
Case-control
Cohort
Cross-sectional
Randomized controlled trial (RCT)
Crossover
Non-inferior trial
What’s a cohort study?
This study type follows the cohort over time (longitudinal) and the outcomes are compared to a subset of the grp who were not exposed to the intervention, such as a drug e.g. Framingham studies
List types of Cohort Study
Prospective
Retrospective (e.g. Through pt medical charts)
What’s unique about randomized controlled trial (RCT)? Effect of this?
Blinding occurs
This may reduce bias
Formula of hazard ratio (HR)?
HR = Hazard in the tx arm/ Hazard in the control arm
What’s single blinding in RCT?
Subjects don’t know what they’re receiving
What’s double blinding in RCT?
Neither subject nor researcher know which subjects received the active drug v. placebo
Uses of RCT?
To test a med for efficacy and safety within a given pt population
What’s pharmacoeconomics?
Pharmacoeconomics research identifies, measures, and compares the costs (direct, indirect and intangible) and consequences (clinical, economic and humanistic) of pharmaceutical pdts and services
List research methods used to determine impact if pharmaceutical pdt or service
Cost-effectiveness
Cost-minimization analysis
Cost-utility analysis
Cost-benefit analysis
Purpose of pharmacoeconomic analysis?
They serve to guide optimal healthcare resource allocation, in a standardized and scientifically based manner
What’s the ECHO model?
Economics, Clinical and Humanistic Outcomes
Uses of ECHO model?
Can be used to determine pharmacoeconomic benefit
What’s the Economic outcomes in the ECHO model?
Include direct, indirect and intangible costs of the drug compared to a medical intervention
What’s the Clinical outcomes in the ECHO model?
Include medical events that occur as a result of the tx or intervention
What’s the Humanistic outcomes in the ECHO model?
Include consequences of the dx or tx as reported by the pt or caregiver (pt satisfaction, quality of life)
List the direct types of costs in pharmacoeconomic.
Direct medical costs
Direct non-medical costs
Indirect costs
Intangible costs
What’s direct medical cost?
Include medications, medication administration, hospitalizations, clinic visit, emergency room visits, nursing services
What’s direct non-medical cost?
Include travel cost (gas, bus, hotel stays fir family), child care services (for children of patients)
What’s indirect cost?
Include loss of productivity of the pt (and possibly of caregiver)
What’s intangible cost?
Include pain and suffering, anxiety, weakness
What’s the main advantage of cost-effectiveness analysis (CEA)?
Outcomes are easier to quantify when compared to other analyses.
Clinicians and decision makers are familiar with these types of outcomes since they are similar to outcomes seen in clinical trials
Which is the most common methodology seen in the literature today?
Cost-effectiveness analysis (CEA)
What’s a disadvantage of CEA?
The inability to directly compare different types of outcomes
E.g. One can’t compare the cost effectiveness of implementing a diabetes program v. Implementing an asthma program where the outcomes units are different (blood glucose values v. Asthma Exacerbations)
When is Cost-Minimization Analysis (CMA) used?
CMA is used when 2 or more interventions have already demonstrated equivalency in outcomes and costs of each intervention are being compared
What’s the limitation of Cost-Minimization Analysis (CMA)?
The use of CMA is limited given its ability to compare ONLY alternatives with the same outcome
What’s Cost-Benefit Analysis (CBA)?
CBA is a systematic process for calculating and comparing benefits and costs of an intervention in terms of monetary units
That is, identifying all the benefits from an intervention and converting them into dollars in the year that they will occur
What’s Cost-Utility Analysis (CUA)?
CUA is a specialized form of cost-effectiveness analysis (CEA) that includes a quality-of-life component associated with morbidity using common health indices such as quality-adjusted life years (QALYs) and disability-adjusted life years (DALYs)
What can you measure using CUA?
Quality of life (years gained) but NOT the “quality” or “utility”
Whats the cost measurement unit of the 4 basic pharmacoeconomic methodologies (CEA, CMA, CBA, CUA)?
Dollars
What’s the outcome unit of CEA?
Natural units (life-years gained, mmHg blood pressure, mg/dL LDL etc)
What’s the outcome unit of CMA?
Assumed to be equivalent in comparative grps
What’s the outcome unit of CBA?
Dollars
What’s the outcome unit of CUA?
Quality-adjusted-life-year (QALY) or other utilities
