Table 3.
Odds ratios from pooled time‐series cross‐section logistic regression analyses predicting high monthly influenza activity levels in American States during the regular flu season, November1999–December 2010 (N = 3712 state‐month observations)*
| Predictor | Widespread influenza | Regional/widespread Influenza | ||
|---|---|---|---|---|
| Odds ratio (95% CI) | Odds ratio (95% CI) | Odds ratio (95% CI) | Odds ratio (95% CI) | |
| Influenza level last month | ||||
| Widespread | 2·620 (1·942, 3·533) | 2·663 (1·978, 3·585) | – | – |
| Regional/widespread | – | – | 3·269 (2·472, 4·322) | 3·268 (2·473, 4·320) |
| Time‐varying state‐month predictors | ||||
| Average influenza level in adjacent states last month | 4·390 (3·473, 5·549) | 4·398 (3·487, 5·547) | 4·411 (3·621, 5·374) | 4·404 (3·620, 5·358) |
| State unemployment rate | 1·199 (1·054, 1·364) | 1·211 (1·071, 1·370) | 1·315 (1·176, 1·471) | 1·302 (1·171, 1·447) |
| State average monthly temp. (°F) | 0·987 (0·960, 1·014) | – | 1·003 (0·980, 1·028) | – |
| State average monthly precipitation (inches, logged) | 0·859 (0·726, 1·017) | – | 0·872 (0·755, 1·007) | – |
| Size of labor force (normed) | 1·149 (0·924, 1·429) | – | 1·105 (0·927, 1·318) | – |
| % metro (normed) | 0·951 (0·842, 1·073) | – | 1·005 (0·907, 1·114) | – |
| Average household size (normed) | 0·968 (0·835, 1·121) | – | 1·006 (0·885, 1·143) | – |
| Average age (normed) | 0·905 (0·755, 1·083) | – | 1·042 (0·889, 1·221) | – |
| % married (normed) | 0·914 (0·802, 1·041) | – | 0·888 (0·793, 0·993) | 0·895 (0·803, 0·998) |
| % college educated (normed) | 0·989 (0·860, 1·137) | – | 1·058 (0·936, 1·195) | – |
| % impoverished (normed) | 0·913 (0·787, 1·060) | – | 0·978 (0·861, 1·111) | – |
| Month fixed effects (ref. = February) | ||||
| January | 1·366 (0·956, 1·951) | 1·341 (0·946, 1·901) | 3·252 (2·279, 4·641) | 3·231 (2·271, 4·597) |
| March | 0·292 (0·205, 0·416) | 0·289 (0·204, 0·409) | 0·298 (0·210, 0·424) | 0·305 (0·216, 0·431) |
| April | 0·023 (0·013, 0·041) | 0·021 (0·013, 0·034) | 0·031 (0·020, 0·050) | 0·032 (0·022, 0·047) |
| May | 0·043 (0·017, 0·109) | 0·033 (0·016, 0·070) | 0·035 (0·017, 0·070) | 0·037 (0·022, 0·061) |
| November | 1·334 (0·618, 2·877) | 1·018 (0·627, 1·653) | 0·758 (0·395, 1·458) | 0·808 (0·537, 1·215) |
| December | 1·456 (0·844, 2·513) | 1·320 (0·877, 1·987) | 1·465 (0·919, 2·336) | 1·555 (1·082, 2·234) |
| Season fixed effects (ref. = 1999–2000) | ||||
| 2000–2001 | 0·703 (0·380, 1·301) | 0·773 (0·429, 1·396) | 1·203 (0·731, 1·978) | 1·222 (0·766, 1·949) |
| 2001–2002 | 1·129 (0·620, 2·057) | 1·192 (0·671, 2·118) | 1·255 (0·768, 2·052) | 1·369 (0·857, 2·185) |
| 2002–2003 | 1·013 (0·530, 1·934) | 1·161 (0·640, 2·107) | 1·245 (0·733, 2·115) | 1·357 (0·835, 1·156) |
| 2003–2004 | 0·802 (0·425, 1·513) | 0·874 (0·496, 1·538) | 0·585 (0·319, 1·072) | 0·667 (0·385, 1·156) |
| 2004–2005 | 2·211 (1·206, 4·056) | 2·366 (1·416, 3·952) | 1·038 (0·604, 1·784) | 1·197 (0·751, 1·907) |
| 2005–2006 | 2·723 (1·399, 5·299) | 3·172 (1·908, 5·276) | 1·176 (0·659, 2·100) | 1·493 (0·942, 2·364) |
| 2006–2007 | 2·183 (1·026, 4·644) | 2·606 (1·577, 4·306) | 1·265 (0·661, 2·422) | 1·660 (1·045, 2·636) |
| 2007–2008 | 3·926 (1·755, 8·781) | 4·806 (2·923, 7·903) | 1·099 (0·544, 2·218) | 1·552 (0·968, 2·489) |
| 2008–2009 | 2·264 (0·870, 5·892) | 2·701 (1·416, 5·152) | 0·899 (0·397, 2·034) | 1·334 (0·758, 2·349) |
| 2009–2010 | 0·129 (0·041, 0·404) | 0·132 (0·054, 0·319) | 0·036 (0·013, 0·096) | 0·051 (0·024, 0·110) |
| 2010–2011 | 0·746 (0·200, 2·785) | 0·691 (0·242, 1·970) | 0·284 (0·097, 0·834) | 0·419 (0·182, 0·961) |
| Likelihood ratio χ2 (d.f.) | 1373·69 (75) | 1362·74 (66) | 2050·05 (75) | 2043·99 (67) |
CI, Confidence interval; d.f., degress of freedom.
*All models include fixed effects for state, month, and flu season.