As falls are an important public health problem, the World Health Organization called for effective intervention strategies in their 2021 report “Step Safely” (World Health Organization 2021). The aim of the present study was to identify characteristics and ICD-10-GM codes in the EHR that enhance identification of patients hospitalized after a fall. We identified demographic information, intrinsic risk factors, type of discharge with related issues and associated ICD-10-GM code categories, including those related to injuries, cognitive deficits and physical limitations. In the upcoming transition to the ICD-11 coding system transferability of the present results to the updated coding ensures their continued relevance and applicability.

In this study, we found that women fell significantly more often than men, which verifies findings of other studies (Bundesärztekammer 2001; Gulich 2004; Nikolaus 2013; Heiß 2015). Several risk factors for falls identified in the present study, including cognitive deficits, incontinence, physical limitations, vascular diseases and other ICD-10-GM codes, were also found in previous studies (Bundesärztekammer 2001; Gulich 2004; Nikolaus 2013; Heiß 2015). Fractures and superficial injuries could be identified as consequences of a fall in this study as well as in previous reports (Gulich 2004; Heiß 2015). In contrast, cognitive deficits and vascular diseases are more probably contributing causes for a fall rather than consequences. Cognitive deficits can cause a lack of orientation, vascular diseases can lead to dizziness or impaired consciousness and physical impairments can lead to an unsteady gait, all of which can contribute to risk of a fall (Arntz 2022).

The group of other ICD-10-GM codes associated with fall prior to hospitalization includes diseases like dermatophytosis and candidiasis that can be linked to a higher need for a nursing environment. These skin diseases could be caused by reduced mobility, which can also increase risk of experiencing a fall. Respiratory disorders and the presence of prosthetic devices or implants (Eveld et al. 2022) may also limit mobility and stability. Symptoms of hypothyroidism and respiratory disorders include fatigue and weakness which could explain the higher prevalence in fallers than among non-fallers. Such considerations as acute posthemorrhagic anemia or a need for immunization against bacterial diseases, such as Clostridium tetani, are likely to represent consequences of a fall and injury. Incontinence can cause falls associated with rushing to the bathroom, especially at night (Nikolaus 2013; Wiedemann 2020).

Our findings include evidence that older patients living with relatives are less likely to be hospitalized due to a fall. However, it remains unclear whether this association reflects better support in daily life or better medical support of the relatives after a fall event as factors tending toward reduce risk of hospitalization due to a fall. The higher number of fallers compared to non-fallers who needed a nursing environment before or after a fall demonstrates the impact on and the need of the health care resources (Behrendt et al. 2021).

Although the intrinsic risk factor “susceptibility to falls” was documented for 2/3 fallers, one third of the non-fallers still had a “susceptibility to falls” documented without a documented fall incident. This difference shows that “susceptibility to falls” does not imply fall incidents in every case.

By analyzing the free texts within the EHR the ICD-10-GM code category X59 contains the only ICD-10-GM codes that were explicitly labeled as “fall” and were therefore associated with a fall incident. At least in the Department of Emergency Medicine of Jena University Hospital, the more frequent documentation of this code in the EHR of fallers than non-fallers makes this ICD-10-GM code important when searching for fallers through EHR.

The ICD-10-GM code categories more often associated with non-fallers evidently represent causes for hospital admission that were not related to a fall. However due to their prevalence in fallers, some medical conditions such as cardiac and pulmonary disorders may or may not be associated with falls.

There are always limitations of generalization of findings from single-site studies. In addition, the quality and accuracy of documentation within EHR are crucial for identifying fall patients. Incomplete or inconsistent documentation may lead to missing or misclassified data, introducing bias and compromising the validity of the findings. Variations in the use of ICD-10-GM codes depend by individual also can lead to variability and unreliability of data. In addition, availability and accessibility of data within the EHR plays an essential role, as only electronically available data were used in this study. In this retrospective analysis, causes versus consequences of fall incidents as well as independent, accompanying events cannot always be clearly separated.