Confidence intervals for difference between Poisson rates


My data consists of occurrences of words in time windows. E.g.:
Day; Word; Frequency
1; "dog"; 45
1; "cat"; 2
2; "dog"; 90
2; "cat"; 4
I would like to estimate the ratio of day-to-day difference (i.e. for dog: 90-45/45 = 100%). For cat the increase is also 100%, but due to the small sample size I would like to somehow quantify that it is "less trustworthy".

Any ideas are most welcome.