Authors’ reply

We would like to thank Furst et al1 for pointing out the potential impact of the healthy vaccine effect on observations made in our recent paper.2 We believe that awareness of such an effect is important for proper interpretation of our results and that the readers should take it into consideration when analyzing our paper.

At the same time, we would like to address several issues related to this phenomenon in terms of our manuscript. We agree that the health literacy of an individual translates into patient outcomes on a population level. It is likely that vaccinated individuals generally take better care of their health, that is, have regular doctors’ appointments, adhere to their medication regimens, participate in screening programs for cancer, are physically active, take better care of their nutrition, etc, which results in lower all-cause mortality. It is probably largely driven by a lower incidence of the most common causes of death (cardiovascular diseases, cancer). Unfortunately, it would be extremely difficult to adjust the analysis for health literacy due to a lack of reliable markers of one’s care of his / her health, especially on a population level. In terms of the health status, we explicitly informed the readers about a lack of adjustment for comorbidities in the Limitations section. Such an approach would probably at least partially limit the impact of the healthy vaccine effect in our study. Interestingly, a series of surveys conducted by the National Institute of Public Health NIH – National Research Institute on adherence to guidelines and restrictions during the initial period of the COVID-19 pandemic before the vaccine rollout showed that elderly people, who tend to have many comorbidities, had a higher health literacy level in the field of COVID-19.3

Additionally, despite some genetic, cultural, and geographical similarities between Poland and Czechia or Hungary, there are several demographic, socioeconomic, and organizational differences which significantly hinder a potential comparison of the countries. One of them is the definition of COVID-19–related death. While in our publication vaccine effectiveness was defined as protection from death directly caused by SARS-CoV-2 infection, COVID-19–related death in Czechia was defined as a death of a person with “a positive COVID-19 polymerase chain reaction test,” according to Furst et al.1 Other differences include those related to vaccination programs; for example, a large part of the Hungarian population was immunized using Sputnik-V and Sinopharm vaccines, which have different effectiveness characteristics, as compared with the vaccines used in Poland.4 It is unclear to what degree such differences modulate the healthy vaccine effect, but they should definitely be taken into consideration. Moreover, we would like to point out that in our study, the vaccination rate increased with age, with an exception of individuals aged 80 years or older. At the same time, increasing age is generally associated with a greater number of comorbidities and higher mortality rate, which seems to be in contrast with the assumptions of the healthy vaccine effect. Due to a lack of data on patients’ medical history, we are unfortunately unable to confirm this hypothesis. Finally, the argument regarding worse access to health care for unvaccinated people in Poland is, to our knowledge, not supported by the literature. The available statistics show that the performance of the Polish health care system improved significantly in 2021, as compared with 2020, thus granting access to appropriate health care for both vaccinated and unvaccinated individuals.5

We are quite surprised by some data showed by Furst et al1; namely, that there was virtually no difference in all-cause mortality among the unvaccinated patients aged 50 to 79 years between the low-COVID and high-COVID periods. One would expect all-cause mortality to be higher in these subgroups during the surge of COVID-19 cases, considering that in the years 2020–2022, COVID-19 deaths accounted for 77% of excess mortality in Czechia.6 We would be very grateful to the authors of the letter for a comment on this interesting observation.