Falls Are Associated With Decreased Autonomy, and Self-Efficacy Moderates This Relation: Results From a National Study

The first aim of this study was to examine the association between falls and perceived autonomy. The second aim was to investigate whether this association is moderated by self-efficacy. Cross-sectional data were drawn from the German Ageing Survey—a nationally representative sample of individuals living in private households aged 40 and above (n = 7,746) in Germany. Perceived autonomy was quantified according to Schwarzer. Self-efficacy was assessed using a widely established scale by Schwarzer and Jerusalem. With covariates being adjusted, linear regressions revealed that experiencing a fall in the past 12 months was associated with lower perceived autonomy (β = −.09, p < .001). General self-efficacy moderated this association (β = .08, p = .02). Findings emphasized the association between falls and perceived autonomy as well as the moderating role of self-efficacy. Future longitudinal studies are required to gain insights into the temporal relationship between these variables.


INTRODUCTION
Experiencing a fall is a frequent phenomenon among individuals aged 65 years and over (1), and the proportion of individuals who experience a fall increases with age (2). Because it is presumed that the number of individuals aged 65 years and over will increase significantly due to demographic changes, and because falls are associated with negative health outcomes, various studies have examined the correlates of falls (3).
Falls can lead to increased depressive symptoms or reduced physical activity (4,5). It has also been shown that falls are associated with loneliness or decreased subjective well-being (6,7). However, it is not known whether falls are associated with perceived autonomy, which is a fundamental and general psychological need (8). Ryan and Deci (8) define autonomy as the "experience of choice". Perceived autonomy leads, for example, to greater psychological well-being (9) and is related to mindfulness (10). Roe et al. (11) found that individuals who did not reflect on previous falls, or did not know why they fell, restricted their lives. However, this study did not explicitly focus on the association between falls and perceived autonomy.
With data drawn from a nationally representative sample of community-dwelling adults aged 40 and older, the aims of this study were 1) to investigate the association between falls and perceived autonomy and 2) to investigate whether the association between falling and perceived autonomy is moderated by self-efficacy.
We hypothesized that falls are associated with a decreased perceived autonomy. Individuals experiencing a fall in the past year may have a lower perceived freedom. More specifically, individuals who experienced a fall may avoid activities of daily living and may avoid leaving the house because they were afraid of falling again (12).
General self-efficacy, the belief in one's capabilities to accomplish tasks or achieve goals, may act as a moderating factor in the relation between falling and perceived autonomy. It appears plausible that individuals who score high in self-efficacy may believe in their ability to achieve goals (e.g., the ability to live autonomously). Thus, self-efficacy might moderate the relationship between falls and perceived autonomy. Or to put it another way: We hypothesized that self-efficacy can moderate the link between falls and autonomy. This is also supported by other empirical studies showing that self-efficacy can act as a coping mechanism [e.g., to buffer against negative effects of informal caregiving (13)]. Furthermore, it has been shown that there is a significant association between falls and (fall-related) self-efficacy (14). However, there is a lack of studies showing that self-efficacy can act as moderating factor in the link between falls and outcome measures such as autonomy.

Sample
We used data from the German Ageing Survey ("Deutscher Alterssurvey, " DEAS). The DEAS study is a longitudinal cohortbased survey of individuals living in private households aged 40 and over (individuals in the second half of life) in Germany.
The current study used the data from the fifth wave (year 2014) because this wave includes measures of autonomy and falls (n = 7,746 individuals provided data on experiencing a fall and perceived autonomy). Neither variable was measured in previous waves. In 2014, the response rate was approximately 61% (panel sample) and about 25% (cross-sectional sample). Neller (15) showed that the response rate is similar to that of other German survey studies. Furthermore, the participation rate mirrors the trend of declining participation rates in Germany. The DEAS study is described in further detail elsewhere (16).

Ethical Statement
Written informed consent was provided by all study participants. The DEAS study follows the principles of the Declaration of Helsinki.
An ethical statement for the DEAS study was deemed unnecessary. Prior to each wave of data collection, the permanent advisory board for the study receives detailed information about the sampling method, the instruments used in the DEAS study, and the consent among respondents to participate. The permanent advisory board concluded that the DEAS study did not require approval from an ethics committee.

Dependent Variables
Perceived autonomy was quantified using a scale developed by Schwarzer (17). It consists of four items (index score from 1 to 4). Items are as follows: "In my daily life, I get along well on my own, " "I make my own decisions and don't allow others to protect me, " "I organize my life according to my own ideas, " and "I cope with my daily life without outside help". The higher the index score, the higher perceived autonomy (Cronbach's alpha = .81).

Independent Variables
The experience of a fall in the past 12 months (yes; no) was measured. This is a common way to measure the history of falls (18,19). Schwarzer and Jerusalem (20, 21) developed a scale to quantify self-efficacy. The original scale consisted of 10 items. The short scale used in this study comprised five items (with four levels each). The short scale was constructed by members of the DEAS study in direct consultation with Ralf Schwarzer. The psychometric properties of the scale have been demonstrated elsewhere (22).
Based on empirical evidence and theoretical considerations, covariates were included. Thus, socioeconomic variables were included in the regression model: sex, age, civil status (single; divorced; widowed; married, living separated from spouse; married, living together with spouse) as well as individual monthly net equivalent income [Organisation for economic co-operation and development (OECD) scale]. With respect to lifestyle variables, the frequency of sports activities and the frequency of alcohol consumptions was recorded (both: "never, " "rarer than once a month," "one to three times a month, " "once a week," "several times a week," and "daily"). Moreover, smoking status was included (daily smoker; casual smoker; former smoker; non-smoker).
Depressive symptoms (15-item version of the Center for Epidemiological Studies Depression Scale, CES-D (23; ranging from 0 to 45; with higher values reflecting more depressive symptoms) and the number of physical illnesses (cardiac and circulatory disorders; bad circulation; joint, bone, spinal, or back problems; respiratory problems, asthma, or shortness of breath; stomach and intestinal problems; cancer; diabetes; gall bladder, liver, or kidney problems; bladder problems; eye problems, vision impairment; ear problems, hearing problems; and other illnesses or health problems) were also included as covariates. In addition, physical functioning was assessed by means of the subscale "physical functioning" of the SF-36 (ranging from 0 = worst score to 100 = best score) (24). Furthermore, selfrated health from 1 = "very good" to 5 = "very bad") served as predicting variable.

Statistical Analysis
Sample characteristics were computed, and pairwise correlations were calculated. With covariates being adjusted, multiple linear regressions were performed to test the relationship between falls in the past 12 months and perceived autonomy and to test whether self-efficacy moderates this relationship. Statistical significance was deemed when the p-value was less than .05. Stata 15.0 (Stata Corp, College Station, TX, USA) was used to conduct statistical analysis.

Data Availability
The data used in this study are third-party data. The anonymized data sets of the DEAS (1996, 2002, 2008, 2011, and 2014) are available for secondary analysis. The data were made available to scientists at universities and research institutes for scientific purposes. The use of the data is subject to written data protection agreements. Microdata (referring to individuals) of all the completed waves of the German Ageing Survey (DEAS) is available free of charge to scientific researchers for non-profitable purposes. The Research Data Centre of the DZA (FDZ-DZA) provides access and support to scholars interested in using DEAS for their research. However, for reasons of data protection, signing a data distribution contract is required before data can be obtained. Please see for further information (data distribution contract) https://www.dza.de/en/fdz/german-ageing-survey/access-to-deasdata.html.

Description of the Sample and Bivariate Correlations
Sample characteristics for individuals providing data for experiencing a fall and perceived autonomy are summarized in Table 1. In sum, 1,355 (17.5%) experienced a fall in the past year. Perceived autonomy was on average 3.5 (± 0.5), and mean selfefficacy was 3.1 (± 0.4).
Pairwise correlations (with Bonferroni-adjusted significance levels) revealed that falls were negatively associated with perceived autonomy (r = −.13, p < .001) and self-efficacy (r = −.10, p < .001) (further details are not shown but available upon request).

Regression Analysis
With several covariates being adjusted ( Table 2), the association between falls and perceived autonomy was examined (first model: first column). This model was extended by adding selfefficacy (second column). Furthermore, an interaction term (falls × self-efficacy) was added (third column).
In the first model, falls were negatively associated with perceived autonomy (β = −.09, p < .001). In the second model, self-efficacy was positively associated with perceived autonomy (β = .39, p < .001). The interaction term (β = .08, p = .02) achieved statistical significance. In other words, the association between falls and autonomy is less pronounced when selfefficacy is high.

DISCUSSION
With data drawn from a nationally representative sample of community-dwelling adults in the second half of life, the purpose of the current study was to investigate the relation between falls and perceived autonomy. Following adjustment for covariates, multiple linear regressions revealed that experiencing a fall in the past 12 months was associated with lower perceived autonomy. This association was moderated by self-efficacy.
We are not aware of any published quantitative research examining the association between falls and perceived autonomy. Moreover, no research has been conducted to date on the moderating role of self-efficacy in the association between falls and perceived autonomy.
The association between falls and decreased perceived autonomy may be explained by the fact that those who have experienced a fall tend to avoid daily activities, such as meeting with friends, because they may fear falling again (12). However, it has been shown that maintaining autonomy can offset the risk of falls among community-dwelling individuals aged 65 or older (25). In practice, this might not always be the case. Future research based on longitudinal data is necessary to clarify the relationship between falls and autonomy.
In our study, it has additionally been shown that the association between falls and autonomy is less pronounced when self-efficacy is high. We assume that individuals who score high in self-efficacy may believe in his or her own abilities to reach the goal of living autonomously. Thus, it appears plausible that selfefficacy moderates this relationship.
The present study has some strengths. To the best of our knowledge, this is the first quantitative study that examines the link between falls and perceived autonomy. Data were drawn from a nationally representative sample of individuals living in private households in the second half of life. Perceived autonomy and general self-efficacy were measured using scales with established psychometric properties. However, some limitations are also worth noting. It has been shown that a small sample selection bias exists in the German Ageing Survey (26). In addition, further information on falls (such as the number of falls or the severity) was not covered in the DEAS study. It is also worth highlighting that this is a cross-sectional study, restricting inference of causality. We cannot dismiss the possibility that our findings are biased by time-constant unobserved factors.  Longitudinal studies are required to overcome this shortcoming. Other factors such as physical functioning or age can also moderate the link between falls and autonomy. Future studies are required to address this issue.

CONCLUSION
In conclusion, findings emphasized the association between falls and perceived autonomy as well as the moderating role of self-efficacy in this relationship. Future longitudinal studies are required to gain insight into the temporal relationship between these variables.

AUTHOR CONTRIBUTIONS
AH and H-HK contributed to the design and concept of analyses, preparation of data, statistical analysis and interpretation of data, and preparation of the manuscript. Both authors critically reviewed the manuscript, provided significant editing of the article, and approved the final manuscript.