Journal Archive

This cross-sectional study investigated the FOF and its influencing factors in older adults with physical independence and in those who underwent surgery after hip fracture.

One hundred older adults who underwent hip fracture surgery were recruited by convenience sampling as orthopedic surgery outpatients from a tertiary hospital in D City, South Korea. This study defined older adults with hip fracture surgery as those who had undergone surgery for hip fracture 6~12 months previously. The control group consisted of 100 physically independent older adults. They were recruited through convenience sampling from a local community senior welfare center that agreed to participate in the study. This study defined older adults with physical independence as those who can perform activities of daily living independently. The inclusion criteria for both groups were as follows: (1) older adults aged ≥70 years; and (2) those who understood the purpose of the study and the content of the questionnaire, were able to communicate, and signed a written informed consent form to participate in the study. The exclusion criteria comprised those who took psychotropic medications owing to issues such as depression. Among physically independent older adults, those with hip fracture surgery were excluded.

To confirm the appropriateness of the sample size, G*Power software version 3.1.9.7. was used. The calculation applied a power of 0.90, significance at p=.05, an effect size of 0.15, and four predictors in linear multiple regression. The minimum sample size required was 99 patients. Therefore, an appropriate sample size (n=100 per group) was used.

We compared the factors influencing FOF in older adults who underwent hip fracture surgery with those who were physically independent. We examined sociodemographic characteristics (sex and age [early old age: 70~74 years; and late old age: ≥75 years]), fall-related characteristics (dizziness and number of falls experienced in the past year), and the Short Falls Efficacy Scale International (FES-I) Korean version (Ko, Lee, & Baek, 2022). The Short FES-I scores were used as the dependent variable.

The Korean version (Ko et al., 2022) of the Short FES-I, originally developed by Kempen et al.(2008), was used to evaluate FOF. The Short FES-I consists of seven items, and participants respond to each item on a 4-point Likert scale (1=not at all concerned, 2=somewhat concerned, 3=quite concerned, and 4=very concerned). The responses to the 7 items were added, ranging from 7 (no concern about falling) to 28 (severe concern about falling). The cut off scores, according to Kempen et al.(2008), were as follows: low concern, 7~8; moderate concern, 9~13; and high concern, 14~28. Cronbach’s ⍺ coefficients during tool development ranged from .93 to .95. The Cronbach’s ⍺ coefficient for this study was .89.

Prior to data collection, a researcher visited a welfare center and an orthopedic surgery outpatient clinic in person and obtained research approval according to the institutional procedures. The data were initially collected from older adults who visited one welfare center and one orthopedic outpatient clinic of a teaching hospital and agreed to participate in the primary study (Ko et al., 2022) to develop of the Korean version of the Short FES-I. This study used the general and fall-related characteristics and the short FES-I data collected from the same data. Data were collected from June to November 2021. Two trained research assistants explained the purpose, methods and procedure of the study and obtained written informed consent from the participants. They also explained the benefits and disadvantages of the research, especially that there are no penalties for withdrawing consent. They read the questionnaire and helped participants respond in-person. It took five to ten minutes per person to complete the questionnaire. The participants in this study were provided with gifts worth KRW 10,000.

Data were analyzed using SPSS/WIN 23.0. The general and fall-related characteristics of the study participants were analyzed using frequency analysis and x² test. The distribution of the short FES-I cut-off scale between the two groups was analyzed using the x² test. Differences in the short FES-I scores according to general and fall-related characteristics between the groups were analyzed using an independent t-test and one-way analysis of variance (ANOVA). Post hoc test of ANOVA used Turkey's HSD. For variables that did not meet the normality test using the Kolmogorov-Smirnov test, the Mann-Whitney and Kruskal-Wallis tests were used for analysis. Multiple linear regression analysis was performed to investigate the factors influencing FOF in each group. The statistical significance level was set at p<.05 for all analyses.

The Institutional Review Board of K Hospital (IRB File No. 2023-03-033) approved this cross-sectional study that used secondary data (Ko et al., 2022) for this research purpose. Before collecting the data, a researcher obtained the research approval in accordance with the institutional procedure by visiting the welfare center and orthopedic surgery outpatient clinic in person. Written informed consent was obtained from the participants after a sufficient explanation of the purpose and methods of the study. Confidentiality and privacy were guaranteed and each data file was coded and anonymized.

In terms of sex, in the group of physically independent older adults (n=100), 17 were male and 83 were female, while in the group of older adults after hip fracture surgery (n=100), 31 were male and 69 were female (x²=5.37, p=. 021). In terms of dizziness, 64 participants in the group with physical independence reported experiencing dizziness, while 42 participants in the group with hip fracture surgery reported experiencing dizziness (x²=9.71, p=.001). In the past year, 31 individuals in the physically independent group had experienced falls once and 17 had experienced falls twice or more. In the group with hip fracture surgery, 75 individuals experienced falls once and 21 individuals experienced falls twice or more (x²=5.82, p<. 001) (Tables 1, 2).

The Short FES-I scores of the two groups were compared by dividing them into low- (7~8), moderate- (9~13), and high-concern (14~28) groups using a cut off scale. According to the cut-off the short FES-I, 11 (5.5%) among the total participants were classified as low concern, 45 (22.5%) were classified as moderate concern, and 144 (72%) were classified as high concern. In the physically independent older adult group, seven individuals (3.5%) were categorized as having low concern, 30 (15.0%) were categorized as having moderate concern, and 63 (31.5%) were categorized as having high concern. In the group of older adults who underwent hip fracture surgery, 4 (2.0%) had low concern, 15 (7.5%) had moderate concern, and 81 (40.5%) had high concern (x²=8.07, p<. 027)(Table 3).

In the physically independent group, the mean Short FES-I scores according to sex were 11.94±4.08 for males and 15.77±4.62 for females (t=-3.04, p=.002). The mean scores according to the presence of dizziness were 13.33±4.56 for “none” and 16.13±4.58 for “present” (t=-2.91, p=.004). The mean scores according to the frequency of falls in the past year were 13.65±3.97 for “none,” 15.35±4.63 for “once,” and 19.18±4.94 for “twice or more”(F=15.21, p<.001).

The average Short FES-I scores in the group with hip fracture surgery were 16.90±5.28 in males and 18.99±4.54 in females (t=-2.01, p=. 047). The Short FES-I score was not significantly associated with older age, dizziness, or number of falls experienced in the past year (Table 4).

As a result of examining multicollinearity before the regression analysis, the tolerance values of the variables were 0.98 to 0.99, which was more than 0.1, the value of the variation inflation factor was 1.01 to 1.02, which did not exceed 10. Therefore, there was no problem of multi-collinearity between the independent variables. The Durbin-Watson value was 1.20 and the autocorrelation between the residuals was small, indicating that the regression model was suitable for testing the independence of the residuals.

To identify factors affecting FOF in the two groups and significant variables, such as sex, presence of dizziness, and recent experience of falls (twice or more), were transformed into dummy variables and included as predictors in the bivariate analysis. The regression model for FOF in the group with physical independence was significant (F=11.28, p<.001), with the most significant predictor being recent experience of falls (two or more times)(β=.35, p<.001), followed by the presence of dizziness (β=.22, p<.05) and female sex (β=.21, p<.05). These three variables explained 26.1% of the variance in FOF. In the group with hip fracture surgery, the regression model for FOF was significant (F=3.01, p<.05), with the variables of dizziness (β=.21, p<.05) and being female (β=.22, p<.05) being significant predictors (Table 5).