Falls among the elderly, risk factors and prevention strategies

Abstract

The occurrence of falls affects approximately one third to one half of seniors over the age of sixty-five and accounts for substantial morbidity, mortality, and disability. Falls that do not result in serious injury, hospitalization or death have the potential to affects seniors' socially and psychologically (e.g., loss of confidence, restriction of mobility, fear of falling) (Kane et al., 1989; Tideiksaar, 1989). Further, despite the low percentage of falls resulting in fractures, the absolute number of seniors that endure fractures taxes the health care system considerably (see, for example, Kellogg International Work Group, 1987). Thus, the provision of accurate information pertaining to the risk factors and preventative strategies for falls would seem essential in times of fiscal restraints.

Unfortunately, progression of knowledge within the area of falls prevention has been hampered primarily by many methodological and conceptual limitations. Additionally, no one data set or research study has been able to adequately deal with all of the issues and gaps that need to be addressed. Therefore, in order to obtain a better understanding of the "big picture" concerning falls, it would seem necessary to tap into multiple sources of data concerning fall information. Secondary data sources, in addition to the collection or primary data, were thus used to address specific gaps within the literature. Specifically, the distinction between one-time fallers and multiple fallers within different settings (e.g., community-based, institutional settings), the existence of effect modification within models, and the analysis of information at the national level pertaining to falls was conducted. Further, the testing of falls education classes and balance control exercise programs within gymnasiums and aquatic settings in the prevention of falls were examined. The data sets utilized for this analysis included the Survey on Ageing and Independence, the National Population Health Survey, data from the Program Needs Survey at Grand River Hospital: Freeport Health Centre, and primary data collected from the completion of the intervention program for falls.

In the institutional setting, similar risk factors were obtained for one-time and multiple fallers. For example, use of psychotherapeutics, experiencing a health change within the last six months, medical diagnosis, impaired mobility and impaired transferring status maintained significance within the final model for time-to-first-fall, while a health change, impaired mobility and transferring status were the risk factors that predicted multiple fall status. Similarly, in the intervention study risk factors between one-time and multiple fallers were quite comparable. Being male, having support of family, a prior history of falling, and impairments in balance/balance confidence were the significant factors in time-to-first-fall. For multiple fallers, hours of sleep, heart conditions, experiencing an external injury within the previous year, and impaired balance/balance confidence were significant.

In the analysis of the Survey of Ageing and Independence several interaction terms maintained significance within the final models. Specifically, interaction terms between age and gender, age and activity limitations, and gender and home maintenance status were obtained within the model for internal injuries. The gender and home maintenance interaction was also obtained within the model for injuries external to the home. These findings suggest that interaction terms are of importance in determining the precise associations in the prediction of falls.

The examination of data at the national level revealed a number of risk factors for falls. General trends between the Survey of Ageing and Independence and the National Population Health Survey showed that being female, advanced age, and measures with respect to social support or homecare services predicted risk of falling; however, several variables that were not common to both data sets were also found to predict risk. For example, medication use and impaired mobility was associated with fall risk in the National Population Health Survey, while homes in need of major repairs predicted risk within the Survey on Ageing and Independence. Therefore, although the two surveys were similar in several respects, the differences that exist offer great insight for future directions in the management of falls among seniors.

The intervention program for falls generally revealed that the seniors participating in the balance control programs experienced significantly less multiple falls, hospitalizations, fractures and deaths than the control group. Further, the experimental group significantly improved their balance and balance confidence after the intervention (Time 2) and at follow-up (Time 3) as compared to the control group. The results from this study warrant the further examination of the balance control programs within a gymnasium or pool setting, as part of a multidimensional risk abatement intervention in the prevention of falls.