Effects of Meteorological Parameters and PM10 on the Incidence of Hand, Foot, and Mouth Disease in Children in China

Abstract

Hand, foot, and mouth disease (HFMD) is a globally-prevalent infectious disease. However, few data are available on prevention measures for HFMD. The purpose of this investigation was to evaluate the impacts of temperature, humidity, and air pollution, particularly levels of particulate matter with an aerodynamic diameter 10 micrometers (PM10), on the incidence of HFMD in a city in Eastern China. Daily morbidity, meteorological, and air pollution data for Ningbo City were collected for the period from January 2012 to December 2014. A total of 86,695 HFMD cases were enrolled in this study. We used a distributed lag nonlinear model (DLNM) with Poisson distribution to analyze the nonlinear lag effects of daily mean temperature, daily humidity, and found significant relationships with the incidence of HFMD; in contrast, PM10 level showed no relationship to the incidence of HFMD. Our findings will facilitate the development of effective preventive measures and early forecasting of HFMD outbreaks.

Keywords: and mouth disease; children; distributed lag nonlinear model; foot; hand.

Figure 1

Map of Ningbo City, Zhejiang Province, China ([17]).

Figure 2

Daily distribution of HFMD incidence and mean temperature, humidity, wind speed, NO₂ level, SO₂ level, and PM₁₀ level in Ningbo.

Figure 3

Relationships among relative risk (RR), temperature, and the number of lag days in Ningbo. After adjusting for relative humidity, wind velocity, wind direction, holidays, seasonal trend, and long-term trends, a three-dimensional plot of RR, temperature, and lag days (n = 10) was generated. Temperature had a nonlinear effect on the incidence of HFMD. We found that the association of temperature with HFMD may have a different lag pattern. For example, the extreme high temperature (31 °C) had a high RR for HFMD cases on the current and fifth lag days.

Figure 4

The RR of HFMD at a specific lag day (0 day,7 day) and at different temperatures (2 °C, 31 °C). At a temperature of 31 °C, there was high RR for HFMD cases. The relationship between RR of HFMD and temperatures had a lag effect. At a temperature of 31 °C, the RR of HFMD had a high result at current day and then decreased for two days and then turned to an increase until the fourth lag day; however, the low temperature (2 °C) had the minimal RR on the current day and had the maximum RR at the second lag day. It considered that the relationship between high temperatures displayed earlier and lasted a longer time than the relationship between low temperatures and HFMD cases.

Figure 5

The overall RR of temperature (°C) for total HFMD cases. The RRs increased with the increment of temperature and it reached the peak at 31 °C.

Figure 6

Relationships among relative risk (RR), humidity, and number of lag days in Ningbo. A three-dimensional plot of RR, humidity, and lag days (n = 10) was generated. Humidity had a nonlinear effect on the incidence of HFMD. We found that the association of humidity with HFMD may have a different lag pattern. For example, the extreme high humidity (90%) had a high RR for HFMD cases on the seventh lag day.

Figure 7

The RR of HFMD by humidity at a specific lag day (0 day, 7 day) and different humidity (35%, 90%). It shows at 90% humidity that there was high RR for HFMD cases. The relationship between the RR of HFMD and humidity had a lag effect. At 90% humidity, the RR of HFMD had a low result on the current day and then turned to an increase until the seventh lag day. When the humidity was 35%, it had a high RR of HFMD cases at the third lag day; however, when the humidity was 90%, the high RR of HFMD cases displayed at the sixth lag day.

Figure 8

The overall RR of humidity for total HFMD cases. The RRs increased with the increment of humidity and it arrived the peak at humidity (90%).

Figure 9

Relationships among RR, PM₁₀, and lag days in Ningbo after adjusting for relative humidity, wind velocity, wind direction, holidays, seasonal trend, and long-term trend. A three-dimensional plot of RR, PM₁₀, and lag days (n = 10) was generated.

Figure 10

The RR of HFMD cases by PM₁₀ at a specific lag days (0 day, 7 day) and the overall relative risk of HFMD according to PM10 level during a 10-day period. It shows no significant correlation was found between the RR of HFMD cases and PM₁₀.