This study tested a large group of Dutch blood donors for IgG antibodies to HEV and examined the risk factors in positive and negative donors. They aimed to find which risk factors were associated with HEV infection in the Netherlands to understand the sources and routes of transmission behind the increasing incidence of autochthonous HEV infection. The study found that seropositivity was unrelated to gender, increased with age, and was associated with consumption of various pork products, particularly dry raw sausages, and exposure to contaminated water (although this route was uncommon). Strengths of the study included using a large, geographically diverse study population; a detailed HEV-specific risk factor questionnaire; and a commonly-used assay that is highly sensitive and specific. Its impact includes confirming the working hypothesis regarding transmission sources of HEV in the Netherlands. This in turn informs precautionary advice to be provided to individuals at risk of chronic infection and may drive and focus further investigations of the contributing sources in the food manufacturing process.

Comments relating to methodology:

1. This study used blood donors as a proxy for the general population (the real population of interest for the study question) but doesn’t specifically state this until the end of the discussion. Ideally this should have been mentioned in the introduction, with more detail in the discussion covering areas such as age, health and socioeconomic status (whether blood donors differ to the general population and how this may impact HEV epidemiology specifically).

2. The authors appear to have used simple rather than stratified random sampling for subject selection. Stratified sampling may have been more valid to ensure the sample represented the general population, rather than the blood donor population, as closely as possible in terms of age, sex and geography. For instance, the study population was approximately 80% male, and although this was consistent between the positive and negative cohorts, it nevertheless biases the findings towards risk factors in men. The authors should either justify the use of simple random sampling or highlight this as a limitation.

3. It is unclear how specific the results are for infection acquired within the Netherlands as the selection criteria did not consider donors’ country of birth or travel history. Given that a similar proportion of positive and negative donors were born in the Netherlands and their travel patterns appear similar, it may be reasonable to assume that most cases were acquired in the Netherlands. Nevertheless, this should be discussed as a limitation. The use of plasma rather than fresh component donors may also have confounded this issue. Certainly for some blood operators, plasma donors would be more likely to have acquired HEV infection overseas because they have a geographical risk that makes them unsuitable for fresh component donation. Usually this risk relates to residence/birth in or travel to areas endemic for malaria, Chagas disease or other arboviruses such as dengue, but there is a significant amount of overlap between these areas and those with high prevalence of HEV. However, it is difficult to assess the impact of using plasma donors without further information on why donors in the Netherlands are directed to plasma donation. The authors should note this as a limitation or justify why it is not.

Minor corrections: 4. The methodology in the abstract could have been clearer by including the use of a multivariate analysis adjusting for age and gender in the methods section rather than the results section. 5. The same statement relating to forcing gender and age into each model is included in both the methods and results sections and could have been left out of the results.