Scarlet fever and meteorological exposures in Jiangsu, China: a time-stratified case-crossover study.

backgroundThere has been an increasing interest in the association between meteorological risk factors and scarlet fever risk. However, the associations between individual-level exposure to meteorological factors and the scarlet fever risk remain poorly understood.

methodsWe collected 36, 912 scarlet fever cases in Jiangsu Province, China (2005–2023) from the Nationwide Notifiable Infectious Diseases Reporting Information System. These data were then paired with daily meteorological factors, including temperature, relative humidity, solar radiation, wind speed, surface pressure, and total precipitation, sourced from the ERA5-Land dataset. A time-stratified case-crossover design was applied using conditional logistic regression combined with distributed lag non-linear models to examine both linear and non-linear associations, while adjusting for public holidays and recent outbreaks. Subgroup analyses by age, gender, period, and season were conducted to assess potential heterogeneity. Effect estimates for individual exposures are expressed as odds ratios (ORs), and interactions were evaluated using the relative excess odds due to interaction (REOI), attributable proportion (AP), and the synergy index (S).

resultsSignificant linear associations (ORs per one-unit increment) were observed for all meteorological variables except wind speed with a 2–5 day lag, peaking at 3 days. Positive associations were found for solar radiation (OR = 1.009, 95% CI: 1.005–1.012) per MJ/[Formula: see text] and surface pressure (OR = 1.088, 95% CI: 1.057–1.120) per kPa, and negative associations for temperature (OR = 0.991, 95% CI: 0.986–0.995) per [Formula: see text]C, relative humidity (OR = 0.995, 95% CI: 0.994–0.996) per %, and total precipitation (OR = 0.994, 95% CI: 0.990–0.997) per mm. All exposures exhibited complex non-linear effects. Temperature showed a reversed U-shaped curve with peak risk at 15–19[Formula: see text]C and fluctuating risk at extreme cold (below −5[Formula: see text]C). Relative humidity displayed an M-shaped curve, peaking at 56–86% and dropping near 39% (OR = 0.97). Solar radiation and surface pressure showed overall declining trends, with elevated risks at lower levels (OR = 1.03 at 1.08 MJ/m[Formula: see text]; OR = 1.01 at 99.82 kPa) and fluctuations across their central ranges. Total precipitation and wind speed showed rising trends, with ORs >1 above 15.1 mm and 5.0 m/s. Stronger associations were observed among individuals aged ≥ 6 years (e.g., temperature OR = 0.988, 95% CI: 0.983–0.994; surface pressure OR = 1.117, 95% CI: 1.077–1.159) and during the post-COVID-19 period (2020–2023), particularly for temperature (OR = 0.989, 95% CI: 0.980–0.999), solar radiation (OR = 1.014, 95% CI: 1.005–1.023), and surface pressure (OR = 1.134, 95% CI: 1.060–1.212).

conclusionsOur study suggests significant associations between meteorological factors and scarlet fever risk, with non-linear exposure-response patterns. We also provide some insights into the mechanisms and targeted prevention strategies for vulnerable individuals, especially in the post-COVID-19 period.