Age-dependent physiological changes, medicines and sex-influenced types of falls

ABSTRACT

Background/Study context: We investigated various parameters related to falls including age-dependent physiological changes, regular medicine use and different types of falls experienced. There is a lack of research investigating the impact of health status, sex, polypharmacy and ageing on different types of falls such as unspecified fall on the same level, mechanical fall on the same level relating to slipping, tripping or loosing balance, fall from a chair, vehicle and fall as a result of syncope, fall from steps or stairs and fall from the height.

Methods: The study included a random sample of 250 older patients, which comprised 10% of the total number of patients (n = 2,492), admitted to a large-scale academic hospital following a fall. Patients’ medicine and illness history, types of falls, liver, renal and sensory function were collected. Univariate analysis was used to examine associations between the type of fall and explanatory variables, followed by multinominal logistic regression analysis.

Results: There was a significant association between the type of fall and sex, p = 0.01, and between the type of fall and regular medicine use, p = 0.002. The multinominal logistic regression analysis revealed that the full model, which considered all explanatory variables together, was statistically significant, p < 0.001. The strongest predictor of all types of falls except ‘fall from the height’ was female sex followed by the regular medicine use.

Conclusion: This study identified predictors for various types of falls in older people; the strongest predictor being a female sex followed by regular medicine use. Based on these findings, the medicine prescribing practice in this older population must be carefully reviewed.

Acknowledgments

The authors gratefully thank the Health Information Services Officers of the hospital who were involved in retrieving medical records to enable data collection for the study.

Conflict of interest

None. The authors declare that they have no conflict of interest.

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.