Chapter 2-CJ Gradoville
2. A-(AR=Number of new cases/number of people at risk)-here the number of people at risk are those who ate both tuna and egg salad.
3. B-Look at the attack rates for the four different cases…for those who at egg salad the attacks rates were 60/75 and 75/100; whereas those who did not eat egg salad the attack rates were 70/200 and 15/50…you can do the same thing for those who did and did not eat tuna. Also compare the attack rate of those who ate egg salad and not tuna (75/100) to those who ate tuna and not egg salad (70/200). The attack rates associated with eating egg salad are much higher than those who ate tuna therefore it is reasonable to conclude the egg salad only is infective.
4. D-Endemic curves can show the type outbreak (direct should show a steady rise (for example venereal diseases probably wont spread like wildfires) while indirect or single source should be more exponential due to its nature and ability to infect a large number of people very quickly). It also can be used to show incubation periods by seeing how much time on the graph it took from introduction to outbreak spike. Herd Immunity is not really associated with outbreaks
5. C-single exposure/common vehicle outbreaks are explosive because they can infect a large number of people in a short period of time.
Chapter 3-CJ Gradoville
1. e-they are measuring prevalence which cannot indicate "risk of developing" which has to do with incidence.
2. 100/1000=10%-since the incidence rate is 5/100,000 people/year and there are 2 milli people in that hood, 2 milli divided by 100,000 =20 so 5x20=100 cases were new that year…therefore 100/1000 who were identified as having schizo were new that year.
3. c-this seems to be the one of the main advantages described in the text
4. c-since the denominator is lowered in each group, all of the incidence rates would increase.
5. d-the data is based on prevalence, not incidence so no conclusions about risk can be made.
6. b-incidence=number of new cases/total pop…so 26/183000=.00014 which is 14/100000.
7. c-prevalence=number of cases/total pop…so 264/183000=.00144 which is 144/100000.
1. 5/1000 - cause-specific mortality rate is # of deaths from a specific disease in a specific period of time divided by the # persons in a population at midyear. So, (30,000/6 million) x 1000 = 5/1000
2. Case fatality rate is the #of individuals dying during a specific period of time after disease onset or diagnosis divided by the # of individuals with the specified disease. So, (30,000/100,000) x 100 = 30%
3. e. Refer to pg. 74 for explanation
4. b. Didn't properly distinguish btwn prevalence and mortality - prevalence is the # of affected persons present in a population at a specific time divided by # persons in pop. @ that time while mortality is when and how the persons died or a measure of the # of deaths in a given population…also mortality is a rate usually on a per year basis while prevalence may only look at the total number of deaths due to some illness up to a particular point in time.
5. b. In this scenario, prevalence rates aren't showing sex differences b/c women are dying from the disease 'more quickly' and aren't being properly represented in the # of affected persons in a population at the certain period
6. d. Proportionate mortality (PM) (not a rate) looks at all deaths and of all the deaths, what proportion was from one disease. Cause-specific mortality rate may be more effective in that we want to look at the death rate by specifying deaths that occurred from a certain disease.
PM may not be as effective b/c if there is a change in the PM from a specific disease over a period of time, the change might be b/c of changes in the mortality of a different disease vs the original disease looked at. (pg 64)
7. d. The standard mortality ratios (SMR) looks at whether people in a certain industry, in this case carpentry, have a higher mortality than people of the same age in the general population.
8. 9.6/1000 b/c (69+115)/(8000+11000) = 184/19,000 = 9.68e-3 x 1000 = 9.68 per 1000
9. e. Proportionate mortality is the # of deaths from Disease Z divided by the ttl deaths (from any disease) *This table doesn't have enough data to answer the question
10. a. Both look at the risk of disease….B/c the disease is fatal and the duration of the disease (survival) is short, mortality is a good index of incidence pg. 67
11. SMR - Observed # of deaths per year divided by Expected # of deaths per year. So, 4,500/1800 = 2.5 x 100 = 250%
Chapter 5-Gucci, aye, okay
1. 1800/2500=72%. I wish I could set up a table but o well. From this info, the number of true pos=1800 and therefore false negatives=700 because there is a total of 2500 who were biopsy proven. The number of false pos=800 and therefore the number of true neg=4200 because a total of 5000 were control and didnt have cancer. Specificity=true +/(true+ (+) false negs) so 1800/2500
2. 4200/5000=84%. See info in 1.
3. 1800/2600=69.2%. Pos Pred value=true+/total pos tests.
4. D-Assuming all other variables are constant (for example it is assumed that the sensitivity and specificity is the same for both populations because its the same screening test), the higher the prevalence the disease the higher the pos pred value is. Since the ratio of false pos to true pos is lowest in A, that means A has a larger pos pred value and thus has a greater prevalence.
5. D-Look at the numbers in the table…pretty self explanatory
6. B-Since Dr. Kidd's test of choice has a smaller sensitivity than Dr. Childs, he will correctly identify less people with strep than dr. Childs. Dont be confused by D because they are testing the same population, the prevalence variable is constant so doesnt matter.
7. 8.4/247=3.4%. Again it would be sweet to draw a table. Since we know the prev is 12/1000 and the sens=70% and spec=75% we can figure out the true pos=8.4 false neg=3.6 false pos=247 and true neg=741. Pos Pred value=true+/total pos tests.
8. B-sequential testing lowers the sensitivity and increases specificity just because
9. 70%-Divide the number they agree on by the total population. They both agreed on 40 abnormal cases and 30 normal cases so 70/100=70%.
10. 57.1%. If you take out the number they both agreed on as normal, this subtracts 30 from the nominator and denominator so it is 40/70=57.1%.
11. Not gonna try to explain this here…maybe later or hope we go over in class.
12. Whatever value we get in 11 if its below .4 it is poor agreement, btw .4-.75 it is intermediate to good agreement and greater than .75 it is excellent.
What is mortality?
-B-killz fo sho kildizzlow