SAS 5: Computing Population Attributable Risk
This assignment uses PROC STDRATE to compute the population attributable risk. We are interested in the relationship between hypertension and stroke. We follow people with and without hypertension for five years and record the number who have a stroke during that period. We have data that looks like this:
Stroke
Yes
No
Hypertension
18
234
No Hypertension
46
952
Note that the second number for each group entered in the data below is the TOTAL for each group and not the number in that group who did not have a stroke.
Use the following program to find the population attributable risk.
DATA stroke ;
INPUT hyper count_h not_hyper count_not;
DATALINES ;
18 252 46 998
;
PROC STDRATE DATA=stroke
REFDATA =stroke
METHOD=INDIRECT(AF)
STAT=RISK
;
POPULATION EVENT=hyper TOTAL =count_h;
REFERENCE EVENT = not_hyper TOTAL = count_not ;
RUN ;
Your assignment:
1. Run the code above.
2. Label the Cumulative Incidence of stroke in patients with hypertension, cumulative incidence of stroke in patients without hypertension and the population attributable risk of stroke due to hypertension
Each statement is explained below.
DATA stroke ;
INPUT hyper count_h not_hyper count_not;
DATALINES ;
18 252 46 998
;
As you know by now, the DATA statement names a dataset being created. I called this stroke since it’s pretty obvious. The INPUT statement has four variables, the number with stroke who had hypertension, the total number with hypertension, the number not hypertensive who had a stroke, and the total population not hypertensive. DATALINES precedes the data, then we have the data itself and a semi-colon to mark the end of the data. Now let’s look at the STDRATE procedure.
PROC STDRATE DATA=stroke
REFDATA =stroke
METHOD=INDIRECT(AF)
STAT=RISK
;
PROC STDRATE DATA = stroke begins the procedure and specifies stroke as the dataset for our study group, in this case, the people with hypertension. REFDATA = stroke specifies stroke as the dataset for our reference population, people without hypertension. The data for both groups happens to be in the same dataset but that is not always the case. From the SAS documentation:
“The AF suboption requests the attributable fraction that measures the excess event rate or risk fraction in the exposed population that is attributable to the exposure. This suboption also requests the population attributable fraction that measures the excess event rate or risk fraction in the total population that is attributable to the exposure.”
STAT = risk specifies that we want risk and not rate. Notice that all of this is one statement.
POPULATION EVENT=hyper TOTAL =count_h;
The POPULATION statement requires two values, the variable giving the number in the exposed group who had the event and the variable giving the total number in the exposed group.
REFERENCE EVENT = not_hyper TOTAL = count_not ;
The REFERENCE statement gives the reference group, usually, the people who were not exposed to the risk factor. This also requires two variables, the number who had the event and the total number in the non-exposed group.
RUN ;
1. You will screen shot your results and put them in a word document for return.
2. You will look for your own 2x2 table for this type of information. Create your own code and explain what it means.
SAS 6: Computing SMR
Standardized Mortality Rate by Strata
This assignment teaches you a little more about reading in data with SAS and also how to compute standardized mortality rates both overall and by strata.
In Part A, you will compute the SMR for skin cancer and compare that to the US as the reference population. In Part B, you will compute the SMR stratified by age. Complete code is given below. Refer to the link in the livebinder for a video explaining each statement.
1. Run the code for the two analyses.
2. Answer this question refe
ing to your results? Does the first analysis show a difference in SMR for skin cancer between Florida and the US as a whole?
3. Answer this question refe
ing to your results? Does the second analysis, by strata, show a difference in SMR for skin cancer between Florida and the US as a whole?
* Assignment 5: PROC STDRATE *
* Part A: Comparing Florida to US *
DATA florida_all ;
event_f = 538 ;
total_f = XXXXXXXXXX ;
DATA us_all ;
event_u = 7420 ;
total_u = XXXXXXXXXX ;
PROC STDRATE DATA=florida_all REFDATA=us_all
METHOD=INDIRECT
STAT=RATE(MULT=100000)
PLOTS=ALL;
POPULATION EVENT=event_f TOTAL=total_f;
REFERENCE EVENT=event_u TOTAL=total_u;
* Part B: Comparing Florida to US *
* By Strata *
DATA florida;
INPUT Age $1-5 Event PYear comma11.;
DATALINES;
00-04 0 953,785
05-14 0 1,997,935
15-24 4 1,885,014
25-34 14 1,957,573
35-44 43 2,356,649
45-54 72 2,088,000
55-64 70 1,548,371
65-74 126 1,447,432
75-84 136 1,087,524
85+ 73 335,944
;
DATA US;
INPUT Age $1-5 Event comma7. PYear comma12.;
DATALINES ;
00-04 0 19,175,798
05-14 1 41,077,577
15-24 41 39,183,891
25-34 186 39,892,024
35-44 626 45,148,527
45-54 1,199 37,677,952
55-64 1,303 24,274,684
65-74 1,637 18,390,986
75-84 1,624 12,361,180
85+ 803 4,239,587
;
PROC STDRATE DATA=Florida REFDATA=US
METHOD=INDIRECT
STAT=RATE(MULT=100000)
PLOTS=ALL
;
POPULATION EVENT=Event total=PYear;
REFERENCE EVENT=Event total=PYear;
STRATA Age / STATS SMR;
RUN;