Syndrome Surveillance for Influenzalike Illness, Appendix

Past Issue

Vol. 10, No. 10
October 2004

Comments


•	Back to article

Research

Syndromic Surveillance for Influenzalike Illness in Ambulatory Care Setting

Benjamin Miller,* Heidi Kassenborg,† William Dunsmuir,‡ Jayne Griffith,* Mansour Hadidi,* James D. Nordin,§ and Richard Danila*
*Minnesota Department of Health, Minneapolis, Minnesota, USA; †Minnesota Department of Agriculture, St. Paul, Minnesota, USA; ‡University of New South Wales, Sydney, Australia; and §HealthPartners Research Foundation, Minneapolis, Minnesota, USA

Appendix

We used this SAS code in fitting the autoregressive model (AUTOREG) that generates the residuals used in the cumulative sum analysis (CUSUM).

PROC AUTOREG DATA=WORK._egtemp_;

/* The time-series identifier, dov, is used in any requested plots. */

MODEL sqrt = Holiday dow cos sin /

NLAG=7

MAXITER=50

METHOD=ML

backstep

DW=13 dwprob

;

OUTPUT OUT=SASUSER.Residuals

LCLM=LCLM UCLM=UCLM PM=PREDICTEDM RM=RESIDUALM

R=RESIDUAL ;

RUN;

The dataset is structured to contain a row for each day in the historical file for all influenzalike illness (ILI) visits in the clinic network. In the code, sqrt is the variable that contains the square root of the count for all ILI visits on day dov. Holiday is a dummy-coded variable for regular or nationally observed holidays, and dow is a dummy-coded variable for weekday or weekend. Sin and cos are variables for seasonal adjustments and can be calculated for each dov by using the respective formulas, where x_n is a continuously increasing integer: or .

proc cusum data = SASUSER.residuals;

xchart residual*dov /

maxpanels = 100

interval = date7

mu0 = –0.0027357 /* target mean for process */

sigma0 = 0.7435702 /* known standard deviation */

delta = 2 /* shift to be detected */

h = 1.1812 /* cusum parameter h */

k = 1 /* cusum parameter k */

scheme = onesided /* one-sided decision interval */

tableall

cinfill = ywh

cframe = bigb

cout = salmon

cconnect = salmon

climits = black

coutfill = bilg;

label residual = ´Cusum of Residuals´;

run;

A dataset is created containing the residuals from the autoregression model and used for the one-sided CUSUM analysis. The residuals are charted for each dov. The example values of delta, h, and k correspond to an average-run-length (ARL) of 50.

Comments to the Authors

Please use the form below to submit correspondence to the authors or contact them at the following address:

Benjamin D. Miller, University of Minnesota Department of Health, 717 Delaware St. SE, Minneapolis, MN 55414, USA; fax: 612-676-5743; email: Benjamin.miller@health.state.mn.us

Comments to the EID Editors

Please contact the EID Editors at eideditor@cdc.gov

		EID Home \| Top of Page \| Ahead-of-Print \| Past Issues \| Suggested Citation \| EID Search \| Contact Us \| Accessibility \| Privacy Policy Notice \| CDC Home \| CDC Search \| Health Topics A-Z

		This page posted August 26, 2004 This page last reviewed September 23, 2004

		Emerging Infectious Diseases Journal National Center for Infectious Diseases Centers for Disease Control and Prevention