Lijun Xu1, Jun Chen2, Zhiying Zhan1, Tingting Chen1, Wenhui Liu1, Jun Qian3, Chunquan Ou4. 1. State Key Laboratory of Organ Failure Research, Department of Biostatistics, Guangdong Provincial Key Laboratory of Tropical Disease Research, School of Public Health, Southern Medical University, Guangzhou, China. 2. School of International Education, Southern Medical University, Guangzhou, China. 3. Department of Mathematics and Physics, School of Biomedical Engineering, Southern Medical University, Guangzhou, China. 4. State Key Laboratory of Organ Failure Research, Department of Biostatistics, Guangdong Provincial Key Laboratory of Tropical Disease Research, School of Public Health, Southern Medical University, Guangzhou, China. Electronic address: ouchunquan@hotmail.com.
Abstract
BACKGROUND: To block transmission of measles, the China Ministry of Health endorsed the "2006-2012 National Measles Elimination Action Plan (NMEAP)" on November 11, 2006, while no published studies used mathematical models to assess the impact of this action. Also, the problem of residuals autocorrelation due to auto-correlated infectious disease data is often ignored in practice. METHODS: Based on monthly counts of measles in 31 Chinese provinces during 2004-2016, a spatiotemporal auto-correlated mixed model was developed to assess the intervention effects over 10 years after the NMEAP, after controlling for seasonality and time-varying socioeconomic and meteorological factors. The first-order autoregression dynamic relationship was used to adjust the bias of conditional covariance in Quasi-likelihood function due to temporal autocorrelation and the spatial autoregressive term was used to capture the neighbor effects of adjacent provinces. RESULTS: There were a similar seasonal pattern and spatial differences of measles incidence among provinces, yet the spatiotemporal pattern was diminished after the intervention. The introduction of the NMEAP was associated with a reduction of 51.03% [excess risk (ER%) = -51.03%, 95% CI: -62.32% to -36.37%] in measles morbidity on average during the 10-year intervention period. The intervention effect varied with time. The significant effect was observed 15 months after the intervention onset, and then gradually increased to a stable level, but was weakened slightly after 2012. The effect of birth rate and meteorological factors on measles morbidity reduced after the NMEAP. CONCLUSIONS: There was clear evidence that prominent progress towards a goal of measles elimination had been achieved after the intervention, while the resurgence after 2012 indicated the goal for measles elimination by 2012 had not been reached.
BACKGROUND: To block transmission of measles, the China Ministry of Health endorsed the "2006-2012 National Measles Elimination Action Plan (NMEAP)" on November 11, 2006, while no published studies used mathematical models to assess the impact of this action. Also, the problem of residuals autocorrelation due to auto-correlated infectious disease data is often ignored in practice. METHODS: Based on monthly counts of measles in 31 Chinese provinces during 2004-2016, a spatiotemporal auto-correlated mixed model was developed to assess the intervention effects over 10 years after the NMEAP, after controlling for seasonality and time-varying socioeconomic and meteorological factors. The first-order autoregression dynamic relationship was used to adjust the bias of conditional covariance in Quasi-likelihood function due to temporal autocorrelation and the spatial autoregressive term was used to capture the neighbor effects of adjacent provinces. RESULTS: There were a similar seasonal pattern and spatial differences of measles incidence among provinces, yet the spatiotemporal pattern was diminished after the intervention. The introduction of the NMEAP was associated with a reduction of 51.03% [excess risk (ER%) = -51.03%, 95% CI: -62.32% to -36.37%] in measles morbidity on average during the 10-year intervention period. The intervention effect varied with time. The significant effect was observed 15 months after the intervention onset, and then gradually increased to a stable level, but was weakened slightly after 2012. The effect of birth rate and meteorological factors on measles morbidity reduced after the NMEAP. CONCLUSIONS: There was clear evidence that prominent progress towards a goal of measles elimination had been achieved after the intervention, while the resurgence after 2012 indicated the goal for measles elimination by 2012 had not been reached.