[The mortality burden of influenza in China: a systematic review]

Objective: To systematically review the mortality burden study of influenza in mainland China. Method: "influenza", "flu", "H1N1", "pandemic", "mortality", "death", "fatality", "burden", "China" and "Chinese" were used as keywords, and a systematic literature search was conducted to identify articles in three English databases (PubMed, Web of Science and Embase) and three Chinese database (CNKI, WanFang and VIP) during 1990-2018 (excluding Hong Kong, Macao and Taiwan). The language of literature was restricted to Chinese and English. The inclusion criteria were human-oriented researches with method based on population, and research indexes included mortality and excess mortality. The exclusion criteria were non-primary research materials, predictive research and research on the burden of avian influenza related deaths. A total of 17 literatures were included, and the basic information to descriptive characteristics, methodology of modeling and the corresponding results were extracted. Results: All the 17 studies adopted indirect statistical models, with 14 of which adopted the regression model, and all the research index was excess mortality. All causes (16 studies), respiratory and circulatory diseases (14 studies) and pneumonia and influenza (10 studies) were the main causes of death associated with influenza. Influenza associated mortality burden in the elderly was higher, with the lowest excess mortality rates of all causes, respiratory and circulatory diseases, pneumonia and influenza being 49.57, 30.80 and 0.69 per 100 000 people, and the highest rates being 228.16, 170.20 and 30.35 per 100 000 people, respectively. In the non-elderly, the corresponding lowest rates were -0.27, -0.08 and 0.04 per 100 000 people respectively, and the highest rates were 3.63, 2.6 and 0.91 per 100 000 people, respectively. The influenza-related excess mortality was higher in the north, with a minimum of 7.8 per 100 000 and a maximum of 18.0 per 100 000, and slightly lower in the south, with a minimum of 6.11 per 100 000 and a maximum of 18.7 per 100 000. There were also differences in deaths caused by different influenza virus subtypes, with influenza A(H3N2) and influenza B virus possibly posing a heavier mortality burden. Conclusions: Studies on influenza mortality burden is mainly based on indirect model and urban level in China. The mortality burden of influenza in the elderly, the northern and subtype A(H3N2) and B were more severe.