Section 5: Future directions (from DOI:10.14218/ERHM.2020.00023)

Navigation

Article	Trends and Prediction in Daily New Cases and Deaths of COVID-19 in the United States: An Internet Search-Interest Based Model
Sections in this Publication
Section	Section 1: Introduction (from DOI:10.14218/ERHM.2020.00023)
Section	Section 2: Methods (from DOI:10.14218/ERHM.2020.00023)
Section	Section 3: Results (from DOI:10.14218/ERHM.2020.00023)
Section	Section 4: Discussion (from DOI:10.14218/ERHM.2020.00023)
Section	Section 5: Future directions (from DOI:10.14218/ERHM.2020.00023)
Section	Section 6: Conclusions (from DOI:10.14218/ERHM.2020.00023)
Section	References (from DOI:10.14218/ERHM.2020.00023)
Named Entities in this Section
Dataset	Pubtator Central BioC-JSON formatted article files

Section 5: Future directions

Despite the high correlation coefficients in retrospective study/modeling, the prediction-models based on the search-interest trend reached low accuracies during a 7-day follow-up. Additional studies are warranted to understand and improve these models. Why the prediction model failed should also be examined. The April 3 CDC recommendation of more indications for mask-use might be one of the reasons. Finally, the factors linked to and the epidemiological significance of lag time revealed by this study should also be further explored.