Looking for the most suitable method for the study of entheseal changes: Application to upper limb's fibrocartilaginous entheses in a human medieval sample

The study of entheseal changes (ECs), that is, osseous changes that occur at muscle attachment sites, has increased in recent years because they have been considered one of the most promising markers of activity in the human skeleton. Their aetiology is highly questioned because several factors unrelated to mechanical stress may be involved in their occurrence. Here, we test the validity of some methods created on different bases for the study of ECs to try to make an approach towards the most beneficial aspects of each method. Seven upper limbs' fibrocartilaginous entheses from a sample of 60 adult medieval individuals have been analysed employing three methods: Villotte (2006), Santana (2011) and the so‐called New Coimbra method (Henderson et al., 2016; Henderson, Mariotti, et al., 2013). These three methods have been applied on the same skeletons. The results show that the three methods seem to indicate the same general trends but with different intensity and some nuances. Santana's method always provides the most frequencies of presence and the highest scores of ECs. In turn, the Coimbra method finds the fewest scores and frequencies of ECs probably because its criteria are the most specific of the three methods. Villotte's method seems to be the one that determines the general tendencies of ECs in the clearest way. This is also evidenced by the calculated Kappa coefficients: the highest agreement is detected between Villotte's lowest scores and Santana's highest ones (k = 0.743); the moderate agreement (k = 0.596; k = 0.544) recorded between Villotte's highest scores and Coimbra's lowest indicates that the latest one registers the lowest scores of ECs between the analysed three methods. All this points to a methodological consensus that should be based on current anatomical knowledge, criteria that are intuitive and flexible, and the use of many scores of osseous expression to avoid excessive sample fragmentation.