Beehler: Biostats Flashcards
What is the frequency/spread of a disease in a population?
prevalence
incidence
attack rate
How well does a test differentiate sick from healthy people?
sensitivity
specificity
*refers to test VALIDITY, how good is this test at comparing those two groups
Of those in a population who test as sick or healthy, how many are truly sick or healthy?
Predictive value
What is the impact of a medicine/treatment?
risk reduction/increase
number needed to treat/harm
What helps you to understand DISEASE BURDEN or the EXTENT of a health problem?
Prevalence (point prevalence)
Period prevalence
during a PERIOD of time
Lifetime prevalence
over the COURSE of a lifeteime
What is the equation for prevalence?
Number of people with a disease at a specific point in time/ Number of people AT RISK for the illness at that point in time
A county in Minnesota has a population of 1,500. In 2013, 180 individuals were diagnosed with type 1 diabetes. Last year, 30 individuals were diagnosed with it. What is the PREVALENCE of type 1 diabetes in this population in 2014?
= (180+30)/1500
= 210/1500
= 0.14
What helps understand the RISK of a specific health event–the number of NEW cases?
Incidence
What is cumulative incidence?
total number reported over time
What is the eqtn for incidence?
Number of NEW people with disease during a time period/ Number of people at risk for illness during that time period
*better for ACUTE cases
A county in Minnesota has a population of 1,500. In 2013, 180 individuals were diagnosed with type 1 diabetes. Last year, 30 individuals were diagnosed with it. What was the INCIDENCE of type 1 diabetes in this population in 2014?
= 30/(1500-180)
= 30/1320
= 0.023
*180 from 2013 are removed b/c they’re no longer new cases
What is attack rate?
Type of incidence used when nature of disease is acute & population observed for short period of time (e.g., outbreaks, specific exposures–flu, chicken pox, exposure)
What is the eqtn for attack rate?
Attack rate = # new cases / # exposed
*short period of time, confined area, exposure is impt (i.e. cholera in Haiti)
What is secondary attack rate?
Secondary attack rate = # new cases / (# exposed – primary cases)
Measures person-to-person spread of disease after initial exposure
*take out ppl who were hit with the flu
Within a kindergarten class, 5 of 35 kids develop chicken pox during a 1-week period. In the next two weeks another 10 kids also come down with chicken pox. What are the attack and secondary attack rates of chicken pox in the classroom?
Attack (whole time)
=(5+10)/35
= 15/35 = 0.43
Sec. Attack (second round)
= 10/(35-5)
= 10/30 = 0.33
What affects prevalence and incidence?
Duration of illness (longer/chronic -> higher prevalence)
Number of new cases (more new cases -> higher prevalence) MORE PPL W/ DISEASE
Migration
- In-migration (ill -> higher prevalence) SICK PPL MOVING IN
- Out-migration (well -> higher prevalence) HLTHY PPL MOVING OUT
- Recovery & death -> lower prevalence
Prevention -> lower incidence REDUCE NEW CASES
Changes in diagnostic criteria or reporting WHAT COUNTS AS SICK/HEALTHY CAN CHANGE (i.e. criteria for AIDS)
What is the relationship between prevalence and incidence?
Prevalence = Incidence x (average) duration* (assumes incidence and duration are stable)
When is prevalence higher than incidence?
disease is long term (diabetes)
When are prevalence and incidence the same?
illness is acute (flu)
What is sensitivity? (test of validity)
SNOUT- negative don’t have it (True positives)
SPIN- positive, do have it (True negatvies)
The probability that a diseased person will be identified correctly by a diagnostic/screening test (AKA true-positive probability or true-positive rate)
What are true positives vs false negatives?
ill identified as ill
ill identified as well
What is the eqtn for sensitivity?
True positives/ Total # ill people [true positives + false negatives]
Highly sensitive tests identify…
most or ALL sick people
A group of individuals who were exposed to Lyme disease were screened using a new test developed for early detection. Of the 344 screened, the disease was confirmed in 258. The new test detected 263 cases of Lyme disease, 32 of which were disconfirmed. What is the sensitivity of the new test?
= True positives / total # ill = 231/258 = 0.90
90% who have Lyme will be correctly identified as having Lyme
What is specificity?
NO N–> TRUE NEGATIVE
The probability that a well HEALTHY person will be identified correctly by a diagnostic/screening test (AKA true-negative probability)
What is the diff between a true negative vs. a false positive?
true neg- well ppl identified as well
false pos- well people identified as ill
What is eqtn for specificity?
= True negative/Total # well people [true negatives + false positives]
What does it mean if you have a highly sensitive test?
High sensitivity -> err on the side of OVER DIAGNOSING
- Identify MOST or ALL possible disease cases
- Most useful when under-diagnosing may lead to severe consequences (e.g., fast developing cancers)
What does it mean if you have a highly specific test?
High specificity -> err on the side of UNDER diagnosing
- Identify most or all well people
- Most useful when over-diagnosing may lead to dangerous, painful, or unnecessary treatment
What happens if you lower the cutoff?
Err on the side of OVER diagnosing
What is predictive value?
how does test work in my population
probability that a test will give the CORRECT dx
What does predictive value depend on?
test sensitivity and specificity
prevalence of the disease in the pop being tested
**vary from pop to pop, study to study
What is the eqtn for positive predictive value?
= True positives/All positives [true + false positives]
*the probability that someone who tests POS for the disease truly has it
= 231/263 = 0.88
What is the eqtn for negative predictive value?
= True negatives/ All negatives [true + false negatives]
*probability that a person who tests NEG truly is healthy
= 54/81 = 0.67
How does high disease prevalence affect predictive value?
Higher positive predictive value (i.e., greater chance that positive test result reflects true illness–SO MUCH DISEASE IN A POP)
Lower negative predictive value (i.e., lower chance that negative test result reflects disease-free status)
How does lower disease prevalence affect predictive value?
Lower positive predictive value (i.e., lower chance that positive test result reflects true illness)
Higher negative predictive value (i.e., greater chance that negative test results reflects disease-free status)
What is a randomized controlled trial?
Have at least one treatment group and one control group
People in both groups may respond positively (e.g., placebo effect) or negatively (e.g., harmful effects)
What is the control event rate?
Proportion of control group participants who have a BAD outcome after “treatment” (e.g., placebo or no treatment)
If 10 of 30 control group participants become sicker after a trial what is the CER?
CER = 10/30 = 0.33 = 33% had adverse outcomes
What is the EER (experimental event rate)?
Proportion of treatment group participants who have a bad outcome after treatment (e.g., new drug)
If 4 of 30 treatment group participants become sicker, what is the EER?
EER = 4/30 = 0.13 = 13% had adverse outcomes
What is absolute risk?
probability of developing a disease or undesired outcome
*tells you whether there’s a reduction or increase as the result of a treatment
CER-EER or EER- CER
What is ARR (risk difference)?
Control event rate is HIGHER than experimental event rate
CER – EER > 0
What is ARI?
Control event rate is LOWER than experimental event rate
CER – EER < 0
After participating in an RCT of a new cancer drug, 10 of 30 control group participants become sicker and 4 of 30 treatment group participants become sicker. Did the new treatment reduce or increase absolute risk? By how much?
CER – EER
= 10/30 – 4/30
= .33 - .13 = .20
= 20% REDUCTION
*As the result of this tx is risk increased or decreased
(relative= control vs. treatment)
What is NNT?
Number of patients who need to be treated to get 1 additional patient a favorable outcome
NNT = 1/ARR
What is NNH?
Number of patients who, if they were treated, would result in 1 additional patient being harmed
NNH = 1/ARI
What is NNT is 5?
For every 5 people treated, 1 more person would respond to the drug
What if NNH is 3?
If 3 were treated, 1 more person would not respond compared with the control group
In a study,200 patients received a new drug to slow the progression of Parkinson’s and 300 patients did not receive the medication. 20 patients in the treatment group and 60 patients in the control group had a typical progression of Parkinson’s. What is the absolute risk reduction and number needed to treat?
CER- EER
60/300-20/200
= .2-.1= 10% REDUCTION
NNT= 1/.1 = 10
