Context Navigation

NM_RateAA100K_GE35.def @ 21905

Last change on this file since 21905 was 21905, checked in by LoisHaggard_NM, 2 years ago
NMBIS Backend QModules - all updates
File size: 12.1 KB

Line
1	#NM_RateAA100K_GE35.def
2	#Calculates directly age-adjusted rates per 10,000 for
3	# records that have been filtered by IBIS-Q (data=tmp)
4	#Uses the IBIS-Q data_frame for cross1 and cross2
5	#Uses an "agepop_frame" so all agegrp18 (agepop elsewhere)
6	# groups are always included/represented (weights will sum to 1.0)
7	#NM_ in filename = applies cell suppression rule
8	#Outputs redflag var based on RSE
9	#
10	f type special
11	#########################################
12	--------BoNdArY--------
13	1 script
14	OPTIONS MPRINT MLOGIC SYMBOLGEN NONUMBER NODATE linesize=175 PAGESIZE=4000;
15	OPTION SPOOL;
16
17	************************ 1. TMP ****************************;
18	* The dataset 'tmp' is the numerator dataset that has been read ;
19	* in already by ibis-q. Any filters have already been applied. ;
20	* The proc summary counts deaths by agepop, cross1, and cross2. ;
21	* The variable "x" must be in the dataset, it is set equal to 1.;
22	****************************************************************;
23	proc summary data=tmp;
24	var x;
25	class agegrp18 %cross1%
26	?cross2? %cross2%
27	;
28	output out=tmp sum=count;
29	proc sort data=tmp; by agegrp18 %cross1%
30	?cross2? %cross2%
31	; run;
32	proc print data=tmp noobs;
33	title1 '---------------------------------';
34	title2 '1. TMP - numerator dataset';
35	run;
36
37	******************** 2. %data_frame% ***********************;
38	* %data_frame% is a dataset created by IBIS-Q. It consists of ;
39	* %cross1% and %cross2% (if the user specified %cross2%) and a ;
40	* variable named "count" that has been set to "0". The results ;
41	* of the proc summary must be merged with the %data_frame% ;
42	* dataset. ;
43	****************************************************************;
44	data agepop_frame;
45	retain count 0;
46	do
47	agegrp18 = 1 to 18 by 1 ;
48	output;
49	end;
50	run;
51	*************************************************************************************;
52	* If user selects specific dimension values, e.g., years, the total dimension value ;
53	* will not be included in the data frame. This creates a dimension value for the ;
54	* total row. But if user does not select specific values (e.g., selects a year, ;
55	* then crosses by sex), there will be a total row for the cross1 and cross2 values. ;
56	* So in a later step, the new_frame dataset is sorted with a "nodupkey" option to ;
57	* remove the additional dimension total values. ;
58	*************************************************************************************;
59	data crosstotalframe;
60	retain count 0;
61	do
62	%cross1%=.;
63	?cross2? %cross2%=.;
64	output;
65	end;
66	run;
67	data df_%cross1%%cross2%;
68	set df_%cross1%%cross2% crosstotalframe;
69	run;
70	proc print data=df_%cross1%%cross2% noobs; title2 'cross1cross2 frame with total row'; run;
71	proc print data=agepop_frame noobs; title2 'agepop frame'; run;
72	proc sql;
73	create table new_frame as
74	select df_%cross1%%cross2%., agepop_frame.
75	from df_%cross1%%cross2%, agepop_frame ;
76	quit;
77	proc sort data=new_frame nodupkey;
78	by %cross1%
79	?cross2? %cross2%
80
81	agegrp18;
82	run;
83	proc print data=new_frame noobs; title2 'new_frame, sorted with nodupkey'; run;
84	proc sort data=tmp; by %cross1% %cross2% agegrp18; run;
85	data newtmp;
86	merge new_frame tmp; *must list frame dataset first, then tmp;
87	by %cross1% %cross2% agegrp18;
88	drop _TYPE_ _FREQ_ ;
89	run;
90
91	proc print data=newtmp noobs;
92	title2 '2. NEWTMP, after frame merged with tmp';
93	run;
94	data tmp;
95	set newtmp;
96	run;
97
98	*********************** 3. POP *****************************;
99	* Sum the pop counts by age group, cross1 and cross2. ;
100	* Read proc summary output, recode the agepop total records, ;
101	* cross1 and cross2 totals to -1. ;
102	****************************************************************;
103	data poptmp; format agegrp18 %popcross1% %popcross2% 8.; set poptmp; run;
104	proc summary data=poptmp;
105	var popcount;
106	class agegrp18 %popcross1% %popcross2%;
107	output out=pop sum=popcount;
108	run;
109	data pop;
110	set pop;
111	drop _TYPE_ _FREQ_;
112	proc print data=pop noobs;
113	title2 '3. POP - denominator dataset';
114	run;
115
116	********************* 4. NUMBERS ***************************;
117	* Join tmp, pop. ;
118	* -1 values are left of out NUMBERS because stdpop doesn't have ;
119	* any -1 values. ;
120	****************************************************************;
121	proc sql;
122	create table numbers as
123	select tmp., pop.
124	from tmp left join pop
125	on tmp.agegrp18=pop.agegrp18
126	?popcross1? and tmp.%cross1%=pop.%popcross1%
127	?popcross2? and tmp.%cross2%=pop.%popcross2%
128	;
129	quit;
130	data numbers;
131	set numbers;
132	if agegrp18^=.; *get rid of total rows to stop doubling from pop proc summary;
133	if %cross1%=. then %cross1%=-1;
134	?cross2? if %cross2%=. then %cross2%=-1;
135	proc print data=numbers noobs;
136	title2 '4. NUMBERS';
137	title3 ' ';
138	run;
139
140	********************** 5. STDWGT ***************************;
141	* Select the appropriate records from the stdwgts datafile. ;
142	* All weights are for U.S. 2000 standard population (per NCHS ;
143	* Statnote20). 'stdvar' values are as follows: ;
144	* agepop - the standard 11 age group weights ;
145	* agepopGE35 - agepop wgts for age>=35 ;
146	* AADist10 - Statnote20 dist #10, for BRFSS ;
147	* AADist10GE40 - AADist10 for ages >=40 ;
148	* AADist10GE50 - AADist10 for ages >=50 ;
149	* AgeGrp18 - 5-year age groups, for cancer data ;
150	* AgeGrp18LT50 - AgeGrp18 for age <50 ;
151	* AgeGrp18GE50 - AgeGrp18 for age >=50 ;
152	* AgeGrp18LT15 - AgeGrp18 for age <15 ;
153	* AgeGrp18LT20 - AgeGrp18 for age <20 ;
154	****************************************************************;
155	data stdwgt;
156	informat stdwgt 7.6;
157	set pop.stdwgts;
158	*pop libname is set in .cfg file with the saspop statement;
159	if (stdvar='AgeGrp18GE25'); *stdvar name is case-sensitive;
160	keep agegrp18 stdwgt;
161	run;
162	proc print data=stdwgt noobs;
163	title2 '5. STDWGT- std. pop. weights for direct age-adjustment';
164	run;
165	****************** 6. NUMBERS2 *****************************;
166	* Adds the standard population weights to the numerator and ;
167	* denominator table. This is done as a separate sql, left join ;
168	* to preserve all the -1 values. ;
169	****************************************************************;
170	proc sql;
171	create table numbers2 as
172	select numbers.*, stdwgt.stdwgt
173	from numbers left join stdwgt
174	on numbers.agegrp18=stdwgt.agegrp18
175	;
176	quit;
177	proc print data=numbers2 noobs;
178	title2 '6. NUMBERS2- Standard POP wgts merged onto dataset';
179	run;
180
181	********************* 7. AGESPECIFIC ***********************;
182	* Calculate age-specific rates & cross-products ;
183	****************************************************************;
184	data agespecific;
185	set numbers2;
186	r=(count/popcount); *age-specific proportion;
187	ratewgt=rstdwgt; weighted age-specific proportion;
188	* Montana method... rate_var=count(stdwgt/popcount)2; New formula from doh.wa.gov;
189	rate_var=(stdwgt*2)((r*(1-r))/popcount);
190
191	proc print data=agespecific noobs;
192	title2 '7. AGESPECIFIC - combines numer and denom data, includes R (age-specific proportion),';
193	title3 'RATEWGT (weighted age-specific rate), and RATE_VAR (rate variance) vars';
194	run;
195
196	********************** 8. AARATE ****************************;
197	* Sum the age-specific rates into aa rates, and merge wgtmax. ;
198	*****************************************************************;
199	proc summary data=agespecific nway;
200	var ratewgt rate_var count popcount;
201	class %cross1% %cross2%;
202	output out=aarate sum(ratewgt rate_var count popcount)=aarate aarate_var count popcount;
203	run;
204	data aarate;
205	set aarate;
206	drop _TYPE_ _FREQ_ ;
207	run;
208	proc sort data=aarate; by %cross1% %cross2%; run;
209	proc print data=aarate noobs;
210	title2 '8. AARATE - summed across weighted agepop cross-products to get AA rate and';
211	title3 'rate variance within each cross-by var';
212	run;
213
214	***************** 9. TMP ***********************************;
215	* Almost final IBIS output dataset. ;
216	* Make sure no summary rows are included. All -1 totals will be ;
217	* coded back to '.' after this step. ;
218	****************************************************************;
219	data tmp;
220	set aarate;
221	if %cross1%=-1 then %cross1%=.; *Recode the statewide totals.;
222	?cross2? if %cross2%=-1 then %cross2%=.;
223
224	if (count>0) then do;
225	rateper=(aarate*100000);
226	rate_se=sqrt(aarate_var);
227	stderr=(rate_se*100000);
228	end;
229	if (count<=0) then do;
230	count=0;
231
232	rateper=0;
233	stderr=sqrt((3/popcount)(1-(3/popcount))/popcount)100000;
234	end;
235	t1=(rateper-(1.96*stderr));
236	if (t1<0) then t1=0;
237	LL=put(t1, 8.2);
238	UL=put((rateper+(1.96*stderr)), 8.2);
239	LL=compress(LL);
240	UL=compress(UL);
241	n=count;
242
243	********************* 10. Red Flag **************************;
244	* redflag is the statistical stability indicator based on the ;
245	* relative standard error (RSE, or coefficient of variation). ;
246	* Redflag values created here are converted to images or special ;
247	* characters in IBIS-View application XSLTfiles, for instance: ;
248	* (xslt\html\query\module\result\ResultPage.xslt, ...Values.xslt ;
249	*****************************************************************;
250	if count>0 then do;
251	rse=(stderr/rateper);
252	redflag=put('-', $14.);
253	if rse>.3 then redflag=put('Unstable', $14.);
254	if rse>.5 then redflag=put('Very Unstable', $14.);
255	if stderr=. then redflag=put('Unstable', $14.);
256	end;
257	if count<=0 then redflag=put('Unstable', $14.); *no variance, n=0, rse=div by zero;
258	run;
259
260	*********** 11. New Mexico Small Numbers Rule *****************;
261	* Suppress cells if the numerator in (1 2 3) AND the denominator ;
262	* is less than 20. For Counts, must run the crude rate code to ;
263	* capture the denominator, but only output the N. ZW's program ;
264	* uses ".A" to identify cells for suppression. I have co-opted ;
265	* his method so I can use the NM logic for cell suppression instead;
266	* of the standard IBIS logic. And I need to use ZW's program ;
267	* because it will suppress the table marginals that can be used to ;
268	* calculate the suppressed cells. If this code is used, the .def ;
269	* file should have the "NM_" prefix. Needs suppressed_variabes ;
270	* code at the end of this file to work. ;
271	*******************************************************************;
272	data tmp;
273	set tmp;
274	if ((0<n<4) and (popcount<5000)) then do;
275	n = .A;
276	rateper = .A;
277	LL = put('**', 8.0);
278	UL = put('**', 8.0);
279	end;
280	proc print data=tmp noobs; title2 '11. TMP - final dataset to pass to IBIS View app';
281	run;
282
283	--------BoNdArY--------
284	f out_variable
285	# definition for output file
286	f xml_out_map_file XMLRate100KNDLCLUCL.map
287	--------BoNdArY--------
288	f out_detail lbl_not_used__see_xml_out_map_file
289	rateper 15.1
290	n 15.0
291	popcount 15.0
292	LL 15.3
293	UL 15.3
294	redflag 15.0
295	--------BoNdArY--------
296	*************** 12. suppressed variables ***********************;
297	* ZW's CGI program must be told how many variables it will need to ;
298	* suppress and which ones they are. NOTE: If the SAS code, above, is;
299	* commented out, these lines can be left in the .def file without ;
300	* causing any problems. They will only be used if the SAS code, ;
301	* above is active, OR if the small_num and small_pop parameters ;
302	* are active in the .CFG file, and with non-zero values. ;
303	********************************************************************;
304	--------BoNdArY--------
305	1 suppressed_variables 2
306	n
307	rateper
308	--------BoNdArY--------

Note: See TracBrowser for help on using the repository browser.

Download in other formats:

Original Format