Time-series datasets of meteorological variables and influenza incidence in Kawasaki City, Japan

In this example, we examined the short‐term associations between mean temperature and influenza cases in Kawasaki City, Japan, from March 2014 to April 2025 using quasi‐Poisson regression combined with distributed lag non‐linear models (DLNMs). These models capture complex non‐linear and multi‐lagged associations by integrating two basis functions: one for the exposure–response relationship and another for the lag–response relationship. Specifically, temperature cross‐basis functions were defined with a natural cubic B-spline having three internal knots at the 10th, 50th and 90th percentiles of the temperature distribution. Lag effects up to 14 days were modelled with a natural cubic B-spline featuring three internal knots equally spaced on the logarithmic scale. Seasonality and long-term trends were controlled for by including a natural cubic B-spline of time with eight degrees of freedom per year. Additional covariates, including a categorical year variable, were included to minimise residual confounding. Two-tailed p-values below 0.05 were considered statistically significant. All analyses were performed in R version 4.1.0 (R Foundation for Statistical Computing, Vienna, Austria).

Data file 1. Daily counts of influenza and meteorological variables in Kawasaki City, Japan, from March 2014 to April 2025. The top panel presents the time series of daily total influenza counts (black line). The second panel depicts the corresponding time series of daily mean temperature (red line).

Data file 2. Association between daily mean temperature and influenza incidence in Kawasaki City, Japan. The top panel shows the RR of total influenza case counts across daily mean temperature and lag dimensions, with 17.3 °C (MMT) as the reference. The bottom-left panel illustrates the RR at cold and hot temperatures (5 °C and 29 °C, respectively) at specific lag days, while the bottom-right panel shows the RR at lag days 4 and 10 across the temperature distribution. Abbreviations: RR, relative risk; MMT, minimum morbidity temperature.

Data file 3. Association between ambient temperature and influenza incidence in Kawasaki City, Japan. Pooled estimates of the overall lag-cumulative exposure–response relationships between daily mean temperature and influenza incidence are presented for: (upper left) total influenza case counts; (upper right) stratification by virus type (A and B); (bottom left) stratification by sex (male and female); and (bottom right) stratification by age groups (0–5 months, 6–11 months, 1 year, 2 years, 3 years, 4 years, 5 years, 6 years, 7 years, 8 years, 9 years, 10–14 years, 15–19 years, 20–29 years, 30–39 years, 40–49 years, 50–59 years, 60–69 years, 70–79 years and ≥ 80 years).