Table 2.

Statistically significant cross-correlation coefficients with Twitter data

Syndrome	Type of data	Time lag (days)	Cross-correlation coefficient
Gastrointestinal	Emergency department visits	− 1	0.26 (0.02–0.50)
ILI	Emergency department visits	− 1	− 0.38 (− 0.62–0.14)
Respiratory	Emergency department visits	− 1	− 0.26 (− 0.50–0.02)
Heat	Telephone health helpline calls	0	0.43 (0.19–0.67)*
Respiratory	Laboratory	1	0.37 (0.12–0.61)
Heat	Temperature	0	0.53 (0.28–0.77)*

*The maximum correlation is provided with the full range of correlations over a range of days included in Supplementary Table 2