We thank Junki Mizumoto for the interest in our study. We acknowledge that the study of Tokuda et al used a more detailed description of the severity of chills.1 The association between shivering and bacteraemia is proven more convincingly by showing that the relative risk increased in more severe categories of chills. However, our data showed no relation between rigors and the outcome sepsis. Rigors corresponds with shaking chills and it is very unlikely that we would have found an association, if we had used the same categorisation used in the study of Tokuda et al. Also, it is important to note that bacteraemia and sepsis are not the same. Shaking chills or rigors might be useful for the decision to prescribe antibiotics, but our study results do not show evidence of added value in the decision to refer a patient to the hospital for (possible) sepsis.
- © British Journal of General Practice 2022
