A cross-sectional online survey compared history of falls, concern about falling and balance confidence between adults with and without chronic ankle symptoms. The study was approved by the institutional human research ethics committee (Approval #: 2014001194) and all participants provided informed consent.
Participants
Study participants were recruited via community advertisements, national and state arthritis organization websites, and social media. Adults with chronic ankle symptoms (symptomatic group) were eligible for inclusion if they experienced ankle joint pain rated as ≥ 2 out of 10 on an 11-point numerical rating scale (NRS) anchored with the ‘No pain’ at 0 and ‘Worst pain imaginable’ at 10 [17], and/or stiffness or reduced movement of ankle in the morning on most days for > 3 months. Participants who had pain in other body regions that was equal to or greater than the ankle pain were excluded. The inclusion criteria for participants with chronic ankle symptoms was based on previous research that found 94% of individuals who have ankle joint pain and/or stiffness for > 3 months have radiographic ankle OA [14]. Participants without chronic ankle symptoms were required to have no history of ankle injury, no ankle pain or stiffness in the past 3 months.
Outcome Measures
Data was collected using an online survey. Participants were asked about their history of falling within the last year, concern about falling and balance confidence. They also provided information on their age, sex, function, comorbid health conditions, ankle pain severity, and pain in bodily locations.
Falls in the last 12 months was determined by the question: “In the last 12 months, have you had any falls?” A fall was defined as ‘an event which results in a person coming to rest inadvertently on the ground, floor or other lower level’ [18]. Participants were categorised as fallers (an individual who fell at least once over the last 12 months) or non-fallers. Participants indicated the number of falls they experienced in the past 12 months from options ranging from ‘0’ to ‘5 or more’. Individuals who selected ‘5 or more’ falls were attributed 5 falls for calculation of total number of falls. Fallers indicated if they sustained an injury from falling, the type of injury experienced (i.e., bruises, cuts/grazes, sprain/strain, broken bones, dislocation), and if any fall resulted in hospitalization.
The Falls Efficacy Scale-International (FES-I) was used to measure concern about the possibility of falling when performing 16 different physical and social activities [19]. Level of concern was measured on a Likert scale ranging from 1 (‘not at all concerned’) to 4 (‘very concerned’). The total (summed) score ranged from 16 to 64, with a higher score indicating a greater concern about falling (i.e., lower falls self-efficacy). Concern about falling was categorised as: low (scores of 16–19),
moderate (scores of 20–27) and high (scores of 28–64) [20]. This instrument has excellent internal and test-retest reliability [19].
The Activities-Specific Balance Confidence (ABC) scale, which has established reliability, was used to measure balance confidence during activities of daily living [21]. Participants were asked to indicate their level of confidence performing 16 activities without losing balance or becoming unsteady. Confidence was rated on a scale from 0% (‘not confident at all’) to 100% (‘completely confident’) with answers provided in 10% increments. An average score was calculated. Scores ranged from 0 to 100%, with higher scores indicating better balance confidence.
The 21-item Activities of Daily Living subscale of the Foot and Ankle Ability Measure (FAAM-ADL) was used to assess function [22]. Each item was scored on a 5-point Likert scale ranging from ‘no difficulty’ (4) to ‘unable to do’ (0). The total score (sum of responses) was converted to a percentage (0-100%), with a higher percentage indicating a higher level of function. This questionnaire has excellent test-retest reliability and internal consistency [22].
Severity of ankle pain was measured using an 11-point NRS. Participants were asked to indicate the worst ankle pain experienced during the past 7 days. Pain in 12 other bodily regions (identified on a body chart) was also recorded using an NRS [23]. The number of bodily regions (including the ankle) with pain ≥ 2 out of 10 were summed as a measure of the number of bodily pain sites (score ranging from 0 to 13).
A modified version of the Self-Administered Comorbidity Questionnaire (SCQ) was used to collect data on co-morbid health conditions [24]. Participants indicated if they experienced and received treatment for 15 medical problems: heart disease, high blood pressure, lung disease, diabetes, ulcer or stomach disease, kidney disease, liver disease, anaemia or other blood disease, cancer, depression, auto-immune disease, back pain, rheumatoid arthritis, gouty arthritis, and osteoarthritis at joints other than ankle. The number of comorbidities were summed (scores ranging from 0–15) as an indication of multi-morbidity.
Data And Statistical Analysis
Survey data was reviewed for completion and participants who did not complete the questions on falls history, concern about falling (FES-I) and balance confidence (ABC scale) were excluded from analysis. Statistical analysis was performed using IBM SPSS Statistics for Windows (Version 25.0. Armonk, NY: IBM Corp).
Chi-square tests were conducted to compare categorical and binary variables (sex, falls status, number of falls, falls-related injuries, falls-related hospitalisation, and concern about falling (FES-I categories)) between participants with chronic ankle problems and controls. A univariate analysis of covariance (ANCOVA) was used to compare differences in concern about falling, balance confidence, and ADL function between participants with and without chronic ankle symptoms. Age, sex, and severity of pain in body in areas other than the ankle were included as covariates. Age was compared between groups using an independent t-test. As data for number of bodily pain sites, severity of pain, and number of comorbid conditions (multi-morbidity) were not normally distributed, Wilcoxon rank-sum tests were used to compare these outcomes between groups. The effect size from the Wilcoxon rank-sum tests was calculated by dividing the Z statistic by the square root of the sample size.
Categorical and binary variables are presented as frequency (percentage), and normally and non-normally distributed continuous variables are presented as mean (standard deviation (SD)) and median (inter-quartile range (IQR)) respectively. For comparison of falls outcomes between groups, risk differences (95% confidence intervals (CI)) were calculated for categorical/binary variables and standardized mean differences (SMD; 95% CI) were calculated for continuous variables. Effect sizes were interpreted as trivial: 0.0-0.2, small: 0.2–0.6, medium: 0.6–1.2, large: 1.2-2.0, very large: 2.0–4.0 and distinct: > 4.0 [25].
The relationship between falls status (dependent variable) and independent variables (age, sex, concern about falling, balance confidence, ADL function, ankle pain severity, pain severity excluding the ankle, number of pain sites, number of morbidities, depression, and group - entered into the model through a dummy coded variable) was investigated using Spearman’s correlation coefficient (normal distribution not assumed). Correlation was interpreted as low: 0.1–0.3, moderate: 0.3–0.5, high: 0.5–0.7 and very high: 0.7–0.9 [25]. The alpha level was set at p ≤ 0.05. A backward elimination regression model was conducted to investigate independent variables associated with falls status (i.e., faller vs non-faller). The model was tested for multicollinearity using the variance inflation factor. Variables were sequentially eliminated leaving only those with p < 0.1.