proc stdrate & failure to display all strata in plot

[lkm]

Good morning:

I am trying to use direct standardization to adjust a target (area of cluster concern) to that of a referent population for 2 variables (maternal age and race/ethnicity). I am trying to figure out why some strata are not displayed in graph below; also I was wondering if there was a way to color-code the graph results to target & referent populations?  The missing strata are consistently for the same age groups (i.e. <20years, 25-29yrs, 35-39yrs) across all race/ethnicities. The data lines are all populating correctly. These missing age group strata appear to be those in between other strata.  I have truncated the data lines to save space. THANK Y'ALL!

**** INPUT COUNTY1 (AREA OF CLUSTER CONCERN) DATA *****;

data County1_999;

County='County1';

input Mom_RaceEthnic $ Mom_Age $ Cases 11. LiveBirths 11.;

datalines;

NH_White  <20yrs       168    5676

NH_White  20-24yrs    568    19692

NH_White  25-29yrs    808    33205

NH_White  30-34yrs    767    29749

NH_White  35-39yrs    354    12174

NH_White  40+yrs       110     2561

NH_Black  <20yrs       164     5020

NH_Black  20-24yrs    364    12040

...;

**** INPUT COUNTY2 (REFERENT/STANDARD POPULATION) DATA *****;

data county2_999_REF;

input Mom_RaceEthnic $ Mom_Age $ Cases 11. LiveBirths 11.;

datalines;

NH_White  <20yrs      75     3589

NH_White  20-24yrs   254   12351

NH_White  25-29yrs   371   21660

NH_White  30-34yrs   420   25120

NH_White  35-39yrs   280   11972

NH_White  40+yrs      75     2834

NH_Black  <20yrs      148   10022

NH_Black  20-24yrs   403   24391

...;

** SORT DATA TO RUN ANALYSIS **;

proc sort data=County1_999;

by Mom_RaceEthnic Mom_Age ;

run;

proc sort data=County2_999_REF;

by Mom_RaceEthnic Mom_Age ;

run;

** RUN DIRECT STANDARDIZATION **;

ods graphics on ;

PROC STDRATE data=County1_999 REFDATA=County2_999_REF

METHOD=direct

STAT=rate(mult=10000)

EFFECT=ratio

;

population group = County event = Cases total= LiveBirths;

reference total=LiveBirths;

STRATA Mom_RaceEthnic Mom_Age/effect;

run;

ods graphics off;

[Reeza]

Show your log please.

[lkm]

NOTE: SAS (r) Proprietary Software 9.4 (TS1M3)

Licensed to TX DEPARTMENT OF STATE HEALTH SERVICES, Site 70206481.

NOTE: This session is executing on the X64_8PRO platform.

　

　

NOTE: Updated analytical products:

SAS/STAT 14.1

SAS/ETS 14.1

SAS/OR 14.1

SAS/IML 14.1

SAS/QC 14.1

NOTE: Additional host information:

X64_8PRO WIN 6.2.9200 Workstation

NOTE: SAS initialization used:

real time 1.56 seconds

cpu time 1.14 seconds

1 ** NAME: DirectStandardCombo.SAS;

2 ** PURPOSE: DIRECT STANDARDIZATION FOR COUNTY CLUSTER**;

3

4

5 title1 "County Cluster Concern ";

6 title2 "Direct Standardization (County2= Referent Population) for Defect:

6 ! Adjusted by Maternal Age & Race/Ethnicity";

7

8 ************************************;

9 ** CREATE DATA FOR STANDARDIZATION **;

10 ************************************;

11

12 **** INPUT COUNTY1 (AREA OF CLUSTER CONCERN) DATA *****;

13 data County1;

14 County='County1';

15 input Mom_RaceEthnic $ Mom_Age $ Cases 11. LiveBirths 11.;

16 datalines;

NOTE: The data set WORK.COUNTY1 has 24 observations and 5 variables.

NOTE: DATA statement used (Total process time):

real time 0.04 seconds

cpu time 0.04 seconds

　

41 ;

42

43 title3 'Area of Geographic Clustering Concern: County1';

44 proc print data=County1;

NOTE: Writing HTML Body file: sashtml.htm

45 run;

NOTE: There were 24 observations read from the data set WORK.COUNTY1.

NOTE: PROCEDURE PRINT used (Total process time):

real time 0.89 seconds

cpu time 0.37 seconds

　

46

47

48 **** INPUT COUNTY2 (REFERENT/STANDARD POPULATION) DATA *****;

49 data County2_REF;

50 County='County2';

51 input Mom_RaceEthnic $ Mom_Age $ Cases 11. LiveBirths 11.;

52 datalines;

NOTE: The data set WORK.COUNTY2_REF has 24 observations and 5 variables.

NOTE: DATA statement used (Total process time):

real time 0.03 seconds

cpu time 0.01 seconds

　

77 ;

78

79 title3 'Standardized/Referent Population: County2';

80 proc print data=County2_REF;

81 run;

NOTE: There were 24 observations read from the data set WORK.COUNTY2_REF.

NOTE: PROCEDURE PRINT used (Total process time):

real time 0.06 seconds

cpu time 0.03 seconds

　

82 title3;

83

84 ************************************;

85 ** SORT DATA TO RUN ANALYSIS **;

86 ************************************;

87

88 proc sort data=County1;

89 by Mom_RaceEthnic Mom_Age ;

90 run;

NOTE: There were 24 observations read from the data set WORK.COUNTY1.

NOTE: The data set WORK.COUNTY1 has 24 observations and 5 variables.

NOTE: PROCEDURE SORT used (Total process time):

real time 0.03 seconds

cpu time 0.01 seconds

　

91

92 proc sort data=County2_REF;

93 by Mom_RaceEthnic Mom_Age ;

94 run;

NOTE: There were 24 observations read from the data set WORK.COUNTY2_REF.

NOTE: The data set WORK.COUNTY2_REF has 24 observations and 5 variables.

NOTE: PROCEDURE SORT used (Total process time):

real time 0.03 seconds

cpu time 0.01 seconds

　

95

96 ************************************;

97 ** RUN DIRECT STANDARDIZATION **;

98 ************************************;

99

100 ods graphics on ;

101 PROC STDRATE data=County1 REFDATA=County2_REF

102 METHOD=direct

103 STAT=rate(mult=10000)

104 EFFECT=ratio

105 ;

106 population group = County event = Cases total= LiveBirths;

107 reference total=LiveBirths;

108 STRATA Mom_RaceEthnic Mom_Age/effect;

109 run;

NOTE: With only one group in the DATA= data set, the GROUP= option in the

POPULATION statement is not used.

NOTE: PROCEDURE STDRATE used (Total process time):

real time 1.26 seconds

cpu time 0.31 seconds

　

110 ods graphics off;

111

112

113 **************************************************;

114 **************************************************;

115 ** END DIRECT STANDARDIZATION **;

116 **************************************************;

117 **************************************************;

[Reeza]

I'm not sure I understand your problem well enough to comment, so if this doesn't make sense, you can ignore it.

I think your strata is age group but you should have a by statement for your ethnic groups, not as part of the strata.

[lkm]

Hi Reeza:

Thank you for your response.

Essentially I am trying to run a multivariable direct standardization. The strata are maternal race/ethnicity * maternal age group:

ex. White <20year

      Black <20year

      Hispanic <20year

      White 20to24year

      Black 20to24year

      Hispanic 20to24year

      White 25to29year

      Black 25to29year

      Hispanic 25to29year

etc

Thank you for your time and insights!

[Reeza]

Create an output data set - does that have all the values in it?

[lkm]

yes

[Reeza · Accepted solution]

If the values are in the output data set but not plot, it sounds like a bug and you need to talk to SAS tech support 😞

[lkm]

thank you