Tracing the source of infection of cystic and alveolar echinococcosis, neglected parasitic infections with long latency: The shaky road of “evidence” gathering

Abstract

We read with interest the work by Torgerson and colleagues “Source attribution of human echinococcosis: a systematic review and meta-analysis” published on June 22, 2020, in PLOS Neglected Tropical Diseases [1]. Echinococcus granulosus sensu lato, causing cystic echinococcosis (CE) and Echinococcus multilocularis, causing alveolar echinococcosis (AE) are transmitted to humans through the fecal-oral route. While identifying and quantifying the role of different routes of transmission to humans would improve control strategies, these information are very difficult to obtain for CE and AE [2]. The results from this systematic review and meta-analysis and from other 2 similar works on the same topic by Possenti and colleagues and Conraths and colleagues, also published in PLOS Neglected Tropical Diseases in 2016 and 2017, respectively [3,4], reach somehow different results. In Possenti and colleagues and Conraths and colleagues [3,4], dog-related factors (e.g., dog ownership, contact with dogs/playing with dogs) were not consistently associated with risk of CE. In the same studies, water-related and food-related factors were not associated with risk of CE, while having a kitchen garden and type of drinking water were associated with the risk of AE [3,4]. Differently, Torgerson and colleagues summarized their results by attributing consistently the majority of infections to “contact with dogs” and “contact with water” for both infections [1]. Interestingly, “living in endemic areas,” which could be interpreted as a proxy for the “environmental contact” of Torgerson and colleagues, was the only factor consistently associated with risk of CE in Possenti and colleagues (association not statistically supported with AE in Conraths and colleagues) but was eventually not analyzed directly by Torgerson and colleagues [1,3,4]. These observations stimulate a due reflection regarding the limits, and, therefore, the usefulness of the systematic review and meta-analysis approach applied to variables that are explored through questionnaires, for long-latent infectious diseases, such as CE and AE, which are usually discovered, if ever, years after the actual infection event has occurred, and that do not cause any acute symptom at the time of infection able to guide the source attribution. While the use of questionnaires is the only way to investigate risk factors for diseases with the abovementioned characteristics, their substantial limits have been largely discussed in 2 recent papers [2,5] and imply that their results should be taken with caution, at least when used to issue practical recommendations on infection control measures. These include, among others, (1) recall bias; (2) bias deriving from the population sampling strategy; (3) difficulties PLOS NEGLECTED TROPICAL DISEASES