Association Among Gender, Age, and Region in Taiwan's First Ten Thousand COVID-19 Cases: A Log-linear-model Analysis (preprint)

Objectives: We explore the association between age, gender, and region among Taiwan's 11290 local Covid-19 cases from January 22, 2020 to June 11, 2021. Methods: Using open data from Taiwan's CDC, we organize them into a three-dimensional contingency table. The groups are gender, age 0-29, 30-59, and 60+ years old, and two classifications for region: (1) 7 commonly-defined regions, (2) 12 groups separating Taipei, New Taipei, Keelung, Taoyuan, Hsinchu county, Miaoli county, and Hsinchu city. We adopt the log-linear model for statistical analysis and use the BIC for model selection. Results: The model with three pairwise interaction terms has the smallest BIC. In terms of interaction effects, there are more females than males among 30-59 (p<0.001), while more males than females among 60+ (p=0.028). Miaoli County has more male than female cases (p<0.001). Differences between 30-59 and 0-29 (baseline), and between 60+ and 0-29 are significant in Taipei (p=0.002 and p <0.001); similar age effects for New Taipei is observed; Miaoli County has significant difference between 60+ and 0-29 (p<0.001). All Taoyuan's interaction terms are not significant. The main effects of age, the differences between 30-59 and 0-29 (baseline), and between 60+ and 0-29, are both significant (p=0.002 and p=0.046). Conclusions: In the four regions with larger numbers of Covid-19 cases, the age and gender characteristics of the infected population are different, reflecting patterns of infection chains.

Taking Account of Asymptomatic Infections in Modeling the Transmission Potential of the COVID-19 Outbreak on the Diamond Princess Cruise Ship (preprint)

BackgroundThe COVID-19 outbreak on the Diamond Princess (DP) cruise ship provided empirical data to study the transmission potential of COVID-19 under quarantine with the presence of asymptomatic cases. MethodsWe studied the changes in R0 on the DP from January 21 to February 19, 2020 based on chain-binomial models under two scenarios: no quarantine assuming a random mixing condition, and quarantine of passengers in cabins -- passengers may get infected either by an infectious case in a shared cabin or by asymptomatic crew who continued to work. ResultsEstimates of R0 at the beginning of the epidemic were 3.27 (95% CI, 3.02-3.54) and 3.78 (95% CI, 3.49-4.09) respectively for serial intervals of 5 and 6 days; and when quarantine started, with the reported asymptomatic ratio 0.505, R0 rose to 4.18 (95%CI, 3.86-4.52) and 4.73 (95%CI, 4.37-5.12) respectively for passengers who might be exposed to the virus due to contacts with asymptomatic crew. The overall R0 for both crew and passengers was decreased to 2.55 (95%CI, 2.36-2.76) and 2.90 (95% CI, 2.67-3.13). Results show that the higher the asymptomatic ratio is, the more infectious contacts would happen. ConclusionsWe find evidence to support a US CDC report that "a high proportion of asymptomatic infections could partially explain the high attack rate among cruise ship passengers and crew." Our study suggests that the effects of quarantine may be limited if the asymptomatic ratio is high, implying that a combination of preventive measures is needed to stop the spread of virus.