We have investigated mid-shaft stress fractures of the bowed femoral shaft (SBFs), well before the first report of an association between suppression of bone turnover and atypical femoral fractures (AFFs). Although all cases of SBF meet the criteria for AFF, SBFs can also occur in patients with no exposure to bone turnover suppression-related drugs (e.g., bisphosphonates). Using bone morphometry and biomechanical analyses, we devised a theory of AFF subtypes, dividing AFFs into fragility SBFs in the mid-shaft and "typical" subtrochanteric AFFs caused by suppressed bone turnover. The aim of this multicenter prospective study was to provide evidence for this novel concept in terms of biological activity. The study was conducted at 12 hospitals in Japan from 2015 through 2019. Thirty-seven elderly women with AFF were included and classified according to location of the fracture into a mid-shaft AFF group (n=18) and a subtrochanteric AFF group (n=19). Patient demographics and clinical characteristics were investigated to compare the two groups. The main focus was on histological analysis of the fracture site, and bone metabolism markers were evaluated to specifically estimate biological activity. All patients in the subtrochanteric AFF group had a history of long-term (>3years) exposure to specific drugs that have been reported to cause AFF, but 5 of the 18 patients in the mid-shaft AFF group had no history of exposure to such drugs. Femoral bowing was significantly greater in the mid-shaft AFF group (p<0.001). In the histological analysis, active bone remodeling or endochondral ossification was observed in the mid-shaft AFF group, whereas no fracture repair-related biological activity was observed in the majority of patients in the subtrochanteric AFF group. Levels of tartrate-resistant acid phosphatase-5b and undercaroxylated osteocalcin were significantly lower in the subtrochanteric AFF group (p<0.05). The possibility of our devised AFF subtype theory was demonstrated. Biological activity tends not to be suppressed in mid-shaft SBFs unlike in "typical" subtrochanteric AFFs involving bone turnover suppression. Although validation of the proposed theory in other populations is needed, we suggest that the pathology and treatment of AFFs be reconsidered based on its subtype.