We read the article by Dunn et al on Ebola infection among health care staff in Sierra Leone with great interest. 1 Health care staff is at a much greater risk of Ebola infection than the population in general. 2 It is unclear how they can be best protected against these risks. 3 Dunn et al provide valuable information on how well personal protective equipment (PPE) works, but they have not quantified this effect. Even though they present their findings as a case study, they can also be analyzed as a retrospective cohort study of how well various levels of PPE use protect against infection in health care workers. The cohort is formed by 64 health care workers who were, ≥1 times, exposed to patients infected with Ebola. They can be divided according to their level of PPE use in no PPE use, gloves only, at least gloves and gowns, or more. Because we know if they became infected or not, we can calculate the relative risks and their 95% confidence intervals (Table 1). We took the total number of exposures as the denominator in calculating the risks as an equivalent of person time. We added 0.5 to prevent cells with zero cases. As can be seen in Table 1, little PPE, such as gloves only, already protects considerably, but more PPE protects better. This was shown in a similar way during the severe acute respiratory syndrome epidemic. 4 This is an important message for educating and training health care staff that have to work in circumstances where fullbody PPE may not always be available. 5