Table 4. Incidence rate ratios of dengue fever and excess risks in 2002 and 2014 explained by time-lagging effect of environmental factors and petrochemical gas explosion event in diverse time periods, Kaohsiung, Taiwan.
| Models | 1–12 months |
6–12 months |
||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| RefYr | 2002 |
2014 |
RefYr | 2002 |
2014 |
|||||
| Ref. | IRR | EREb | IRR | EREb | Ref. | IRR | EREb | IRR | EREb | |
| Base model (BM)a | 1.0 | 11.2* | 34.5* | 1.0 | 11.5* | 35.3* | ||||
| Model 1: BM + MinT (i-month lag, positive effect) | ||||||||||
| MinT: 1-m lag + 2-m lag + 3-m lag + 2-m lag x 3-m lag | 1.0 | 9.4* | 18.0% | 20.1* | 42.9% | 1.0 | 9.6* | 18.0% | 20.5* | 43.1% |
| Model 2: Model 1 + Rainfall (i-month lag, negative effect) | ||||||||||
| Rainfall: 1-m lag + 2-m lag + 1-m lag x 2-m lag | 1.0 | 7.9* | 32.4% | 19.7* | 44.1% | 1.0 | 7.9* | 34.1% | 18.9* | 47.7% |
| Model 3: Model 2 + PGE | ||||||||||
| PGE: occurred on July 31, 2014 | 1.0 | 7.8* | — | 9.3* | 75.1% | 1.0 | 8.0* | — | 9.1* | 76.5% |
Note: *P < 0.05; IRR: incidence rate ratio; MinT: minimum temperature; PGE: petrochemical gas explosion; RefYr: reference years (the period during 2000 to 2014, but beyond 2002 and 2014; a total of 13 years).
aBase model measured the IRRs of 2002 and 2014 as compared to the reference years.
bExcess risks explained (ERE) by covariates additionally added to the study model.