Volume 19, Number 8—August 2013
Research
Emergency Department Visit Data for Rapid Detection and Monitoring of Norovirus Activity, United States
, Sherry Burrer, Soyoun Park, Tarak Trivedi, Umesh D. Parashar, and Benjamin A. Lopman
Table 1
Linear regression model estimates of the association between norovirus outbreaks and BioSense emergency department visit data, by age group, United States, January 2007–April 2010*
| Subsyndrome and age group, y | Norovirus,† β1, × 10−4 (95% CI) | p value‡ | R |
|---|---|---|---|
| Diarrhea | |||
| 0–4§ | 1.72 (1.15–2.29) | <0.0001 | 0.926 |
| 5–17¶ | 1.59 (1.23–1.95) | <0.0001 | 0.828 |
| 18–64¶ | 0.70 (0.56–0.83) | <0.0001 | 0.864 |
| >65§¶ | 0.71 (0.55–0.86) | <0.0001 | 0.917 |
| All ages§ | 0.94 (0.74–1.14) | <0.0001 | 0.910 |
| Nausea/vomiting | |||
| 0–4§ | 7.01 (4.93–9.09) | <0.0001 | 0.866 |
| 5–17 | 5.73 (4.24–7.23) | <0.0001 | 0.796 |
| 18–64¶ | 1.60 (1.14–2.06) | <0.0001 | 0.758 |
| >65 | 0.61 (0.31–0.91) | 0.0002 | 0.729 |
| All ages | 2.78 (2.15–3.40) | <0.0001 | 0.832 |
*Emergency department chief complaint–based subsyndrome visits as a monthly proportion of all visits in 6 states (GA, MO, OH, PA, TN, WY) regressed on norovirus and rotavirus surveillance data and time variable. Intercept for each model p<0.0001.
†Suspected and confirmed norovirus outbreaks.
‡By t test.
§Included term for proportion of rotavirus tests positive in the model, as it is significant and positive (p<0.05). β2 for 0–4-y age group diarrhea model = 0.0539 (95% CI 0.0403–0.0674); β2 for 0–4-y age group nausea/vomiting model = 0.0745 (95% CI 0.0252–0.1237); β2 for ≥65-y age group diarrhea model = 0.0048 (95% CI 0.0011–0.0084); β2 for all-ages age group diarrhea model = 0.0065 (95% CI 0.0018–0.0112).
¶Time variable not significant in model (p>0.05).