Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
44 Cards in this Set
- Front
- Back
Morbidity |
Any departure from a state of well-being |
|
Mortality |
You dead nigga |
|
Two ways of Measuring Risk |
-Probabilities -Rates |
|
Probability |
Proportion of people unaffected at the beginning of study period, but who experience a risk event during the study period |
|
Rate |
Number of events that occur in a defined time period, divided by average number of people at risk (usually population of people at middle of period) Rate = (Persons with disease/Persons at risk) x c |
|
Rate is a good approximation of risk if |
-Event in numerator occurs only once during study period -Proportion of population affected by event is small -Time interval is relatively short |
|
Measures of Morbidity |
-Incidence rate -Prevalence "rate" |
|
Incidence Rate |
(# of new cases / population at risk) x k Measures risk of developing a disease |
|
Prevalence "Rate" |
#Active Cases / Population at Risk Measures proportion of persons with disease Two types: period prevalence, point prevalence |
|
Measures of Mortality |
-Mortality Rate -Case-fatality rate -Proportionate mortality ratio |
|
(Crude) Mortality Rate |
#deaths/population Measures risk of dying |
|
Cause-Specific Mortality Rate |
# deaths by disease / population Measures risk of dying from a disease |
|
Case-fatality Rate |
# deaths by disease / population with disease Measure of lethality |
|
Survival Rate |
1 - (case-fatality rate) |
|
Proportionate Mortality Ratio |
# deaths from one cause / all deaths Not a measure of risk, measures relative importance of a specific cause of death relative to all deaths. |
|
Direct Standardization of Rates |
Use an external standard population to calculate expected number of deaths with ACTUAL rates of mortality. This is done to control for a specific variable. |
|
Standardized Mortality Ratio |
Used when rates are not available for given observed values. Similar process as standardization of rates. |
|
Relative Risk |
Risk with factor / Risk without factor Measures strength of association between exposure and disease. |
|
Attributable Risk |
Risk with factor - Risk without factor Measures the amount of absolute risk that is attributable to factor |
|
Attributable Risk Percent (Attributable Fraction) |
[Risk (Factor) - Risk (No Factor)] / Risk (Factor) |
|
Population Attributable Risk
|
[Risk(pop)-Risk(no factor)]/Risk(pop) |
|
Study Validity |
When results of study are true and meaningful |
|
Internal Validity |
When results of study are true and meaningful for patricipants |
|
External Validity |
Generalizability of study When results of study are true and meaningful for a larger population, not just participants |
|
Random Sampling |
Selecting small group (n) for study from larger population (N) by use of a random method (R) |
|
Simple Random Sampling |
-Assigning unique number to possible elements from N -Selecting a small group (n) using a random procedure |
|
Systematic Random Sampling |
-Assign unique number to elements of N -Order from smallest to highest -Calculate N/n=k -Use random method to select starting element -Select elements sequentially every Kth element from N (Never use starting point as first element in n). |
|
Stratified Random Sampling |
-Organize your elements from N into homogeneous sub-populations ( strata) -Perform simple or systematic random sampling on each strata to obtain your final population n. -Helps avoid sampling errors |
|
Cluster Random Sampling |
-Identify groups (clusters) of elements -Assign unique number to each cluster -Use random method to select cluster -Select all element from chosen clusters |
|
Standard Error of the Mean (SEM) |
Standard deviation of a Sampling Distribution SEM = sd/sqrt(n) |
|
95% Confidence Interval |
Sampling Mean +/- 1.96 (SEM) |
|
As sample size increases, the confidence interval |
Gets narrower, SEM gets smaller |
|
As sample size decreases, the confidence interval |
Gets wider, SEM gets larger |
|
Type I error (alpha) |
Rejecting a Null Hypothesis when it is actually true. Same as p-value. |
|
Type II error (Beta) |
Accepting a Null Hypothesis when it is actually false. |
|
Statistical Power |
Ability of a test to reject null hypothesis when it is actually false. Probability of not committing a Type II Error Statistical Power = 1 - Beta |
|
Acceptable levels of significance and power |
Significance: alpha = 0.05 Power: 1 - Beta = 0.80 |
|
z-test Student's t-test |
Compare means of two groups Requires a normal distribution (t-test) |
|
Analysis of Variance (ANOVA) |
Compare means of three or more groups Requires a normal distribution |
|
Chi-Squared |
Compare proportions of two or more groups |
|
Fisher's exact test |
Compare proportions of two groups (if small "expected frequencies") |
|
Post-Hoc Analysis |
Situation in which you choose which comparisons to make AFTER you've looked at the data
|
|
Mann-Whitney test |
Non-parametric test to compare two groups that do not have a normal distribution |
|
Kruskal-Wallis |
Non-parametric test to compare three or more groups that do not have a normal distribution |