Application of Richards model in real-time incidence prediction of pandemic influenza A (H1N1)

Objective To study the characteristics of pandemic influenza A (H1N1) in Zhejiang province. Methods Richards model was fitted by using the incidence data of pandemic influenza A (H1N1) in Zhejiang during 2009 - 2010 to estimate the inflection point of epidemic, basic reproduction number and total case number. Results A total of 13 487 pandemic influenza A (H1N1) cases were reported from May, 2009 to March, 2010 in Zhejiang with the incidence of 26.04/lakh. There were two incidence peaks, the first one was on September 22, 2009, the basic reproduction number was 2.44 (95%CI:1.75-3.96). The incidence rose again on October 10, 2009 and peaked on November 28, 2009, the second peak could be found 16 days later. The estimated case number was 13 642.82 and the actual case number was 13 706, the error rate was only 0.46%. The coefficient of determination (R square) for Richards model was 0.999 and the basic reproduction number was 1.33 (95%CI:1.31-1.35). Conclusion Richards model is effective in real-time incidence prediction of pandemic influenza A (H1N1) and could found the inflection point of epidemic in time.