Epidemiology Flashcards
Event rate in control group
Event rate experimental group
Absolute risk
Occurs because screening test are more likely to find slow growing tumors rather than those that are rapidly growing. This can bias in favor of screening because more slow growing cancers with a good prognosis will be found with a screening test
Lead time bias
Occurs because study volunteers and persons willing to undergo screening test are likely to be compliant. Compliant patients have better outcomes
Compliance bias
Occurs when two or more factors are associated with the outcome and only the one being studied is accounted for
Confounding bias
Occurs when subjects for the study do not represent the population. This is avoided by having large representative samples
Selection bias
This occurs when care provided to the experimental and the control group differ substantially other than the intervention. It is avoided by blinding. Patients and doctors tend to behave differently when they know they are being studied. Radiomized blinded trials protect against this bias
Performance bias
There is no consistent relationship between blood levels of morphine and analgesic effects. Thus no single right dose. You should titrate to desired effect. This is because of?
Tolerance
Hypothesis that a there is no significant difference between two groups being compared to each other Thus is the p value is less than 0.05 REJECT IT
Null hypothesis
When error occurs because too few are tested
Type 2 error
Calculate the median number
127,104,103, 81, 86, 117, 89, 97, 95
First rank the numbers: 81, 86,89, 95, 97, 103, 104, 117, 127 then add the total number +1 divided by 2 to identify
Median= 97
Calculate the mean?
81,86,89,95,97,103, 104,127,127
N=the total number included in data set n=9 81+86+89+95+97+103+104+117+127= 899 =899/n = 899/9 =99.89
Calculate mode
81, 86, 89, 95, 97, 103, 104, 117, 127
The mode of a sample of data is the observation that occurs most frequently
There is no unique mode since each number in data set is listed only once
The more cases you detect , the more false negatives will be included in your sample. Thus best screening test includes both both high sensitivity and high specificity
As sensivity increases specificity decreases
In study diabetic pts w no intense treatment had 10% neuropathy, diabetic pts with intense therapy had 2% neuropathy how many diabetic patients need to be treated with intense therapy to prevent neuropathy?
first test is for NTT
NTT= 1/ABSOLUTE RELATIVE RISK ARR= control group event rate - experimental group event rate ARR= 10-2%= 8% = .08 NTT=1/.08 =12.5
To prevent a disease from occurring at all by removing it’s cause
Primary prevention
To detect disease early so intervention can prevent progression
Secondary prevention
To reduce complications and mortality due to disease after it is recognized
Tertiary prevention
Study of select subjects who are categorized based on outcomes and try to find associations with certain variables. Does not follow patients over Time
Case control
Cohort looks at groups from zero and follows for specific time to find association the variable is not under researcher control
Cohort study
Statistical method that allows for adjustment of con founders
Multivariate analysis
A systematic review that combines the results of previous studies to evaluate the magnitude or direction of an effect on a subgroup
Meta analysis
How often test will pick up,disease if it is there
Sensitivity
The proportion of pts who do not have the disease and who will test negative for it
Specificity
The amount of times the intervention or exposure increases (decreases) the risk of the outcome
The ratio incidence of outcome among exposed persons divided by total number
Relative risk.
“. Disease X Absent
Test A result positive 30 (a) TP 50 (b) FP
Test A result negative 10 (c) FN 80 (d) TN”
Excerpt From: Alfred F. Tallia, Joseph E. Scherger & Nancy Dickey. “Swanson’s Family Medicine Review.” Saunders, 2013. iBooks.
This material may be protected by copyright.
Check out this book on the iBooks Store: https://itun.es/us/jwxmI.l
N
