Sigmoids are challenging to fit. In this case I would guess the biggest problem would be that the upper plateau parameter is almost not constrained by data at all… (this is also problem for any predictions from such a model). Finally the data is messy and has all sorts of reporting biases and non-stationarity due to evolving reactions of the governments and public, so a sigmoid might be a problematic model.
I’ve found the direct parametrization of the sigmoid problematic. I worked with @stemangiola on fitting sigmoids in a different context, you might want to check out the parametrization we used in a preprint: https://doi.org/10.1101/2020.03.16.993162, also discussed at Difficulties with logistic population growth model
With epidemic modelling in this time I would also ask you to be very careful about communicating the results. There is a lot of noise in the public communication channels that can easily drown out the voices of the best experts so please make sure you are not adding to this noise.
Best of luck with your model