Older People's Views of Falls-Prevention Interventions in Six European Countries

Abstract

Reducing the incidence of falls among older people is a public health priority. More than one third of people over the age of 65 years fall each year, and the incidence is higher among the very old and frail (Speechley & Tinetti, 1991). However, serious falls also are incurred by fit and active older people (Allander et al., 1998). Falls are consequently the most common cause of accidental injury among older people, which in turn can lead to permanent loss of function and even death (American Geriatrics Society, British Geriatrics Society, & American Academy of Orthopaedic Surgeons Panel on Falls Prevention, 2001; Todd et al., 1995). In addition, falling and fear of falling are linked to elevated levels of psychological distress, restriction of activity and independence, and increased social isolation and use of health and social services (Bruce, Devine & Prince, 2002; Cumming, Salkeld, Thomas, & Szonyi, 2000; Delbaere, Crombez, Vanderstraeten, Willems, & Cambier, 2004; Howland et al., 1998; Murphy, Williams, & Gill, 2002; Yardley & Smith, 2003).

There is a growing body of evidence indicating that falls-prevention programs that include muscle strengthening and balance-training exercises can significantly reduce the incidence of falls (American Geriatrics Society et al., 2001; Chang et al., 2004; Gillespie et al., 2001; Skelton & Todd, 2004). However, although the efficacy of these programs has been demonstrated, their effectiveness at reducing the rates of falling in the population depends crucially on rates of uptake and adherence. Although high rates of participation and adherence have been achieved (Barnett, Smith, Lord, Williams, & Baumand, 2003; Tinetti et al., 1994), typically fewer than half of those invited to take part in falls-prevention interventions in the community take up the opportunity (Campbell et al., 1997; Robertson, Devlin, Gardner, & Campbell, 2001; Stevens, Holman, Bennett, & de Klerk, 2001). Uptake can be as low as 10% (Day et al., 2002; Fabacher et al., 1994), and there is then further attrition through dropout and nonadherence. It is therefore important to improve our understanding of how prevention programs can be designed and presented so as to maximize acceptability and participation among older people.

Because exercises form a key component of successful interventions, the literature on factors motivating older people to undertake generic exercise and vigorous physical activity may prove relevant. Research suggests that most older people prefer to exercise at home, but with some professional guidance (Hillsdon, Thorogood, Anstiss, & Morris, 1995; King et al., 2000). Uptake and adherence to interventions to encourage older adults to exercise is associated with a history of being physically active, lower levels of illness, greater self-efficacy (i.e., the belief that one is capable of exercising), and a perception that exercise improves well-being (Martin & Sinden, 2001). Barriers to exercising include pain, illness, and fear of causing physical harm (Martin & Sinden; O'Brien Cousins, 2000; Resnick & Spellbring, 2000; Stead, Wimbush, Eadie, & Teer, 1997), and low levels of social approval and support for exercising and vigorous physical activity in later life (Brawley, Rejeski, & King, 2003; O'Brien Cousins & Janzen, 1998; Stead et al.).

Predictors of uptake and adherence to interventions to reduce falls risk may differ from predictors of generic exercise for several reasons. In falls-prevention interventions very specific physical activities are prescribed, often alongside other lifestyle and medical interventions, for the particular purpose of reducing risk of falling. Reported participation rates vary widely in published clinical trials, and this may be because of the content of the interventions (e.g., hazard reduction vs exercise), the format of the intervention (e.g., group vs home based), how participation is encouraged (e.g., community action vs health professional prescription), or how the population is sampled (e.g., unselected vs high risk). For example, after having a home-based assessment, 87% of older people followed advice to see their doctors about a medical risk and 71% followed advice to modify their homes to reduce hazards, but only 54% increased their activity as recommended and approximately 33% (one in three people) followed advice to avoid drinking alcohol (Fabacher et al., 1994). In a multifactorial intervention among older people with cognitive impairment, nearly two thirds of the individuals adhered to strength and balance-training exercises, whereas only half of them modified their medications and less than half (39%) reduced home hazards (Shaw et al., 2003). It is therefore difficult to identify what factors may have been responsible for varying rates of uptake and adherence in published interventions, because they differ on a number of potentially relevant dimensions.

Qualitative studies carried out in the context of a variety of falls injury-prevention programs can help to explain the reasons for varying rates of participation in different types of interventions. Frail older people interviewed as hospital inpatients were often unaware that any form of exercise could help prevent falls, and they viewed exercise as too vigorous an activity for them to undertake (Simpson, Darwin, & Marsh, 2003). Conversely, older adults in the community can be hostile to the idea of falls prevention chiefly because they assume that this will involve restricting their activity (Commonwealth Department of Health and Aged Care, 2001; Yardley & Todd, 2005). Home-hazard reduction is sometimes seen as an intrusive interference into personal choices about lifestyle (Clemson, Cusick, & Fozzard, 1999; Simpson et al.), while use of hip protectors is influenced by concerns about comfort, convenience, and appearance (Cameron & Quine, 1994).

Although these studies have provided useful insights into motivations and concerns regarding participation in falls-related interventions, the context-sensitive nature of qualitative research means that it is problematic to generalize their findings beyond the particular setting of each study. Moreover, these studies not only investigated a range of different interventions in different populations, but they also used different interview questions and methods of analysis, which makes it more difficult to synthesize their findings (McInnes & Askie, 2004). Therefore, our aim in the present study was to identify factors promoting or inhibiting uptake and adherence that are common to a wide range of contexts, including different geographical and cultural settings, different community-living populations, and different experiences of falls-related interventions.

To achieve this aim, we used the same semistructured interview schedule in each of the diverse contexts we sampled, asking all participants about interventions that included strength-and-balance training, because this is a crucial component of successful falls prevention. Our interview questions were based on the theory of planned behavior (Ajzen, 1991), a well-validated model that proposes that an individual's intentions and behavior are predicted by beliefs about the positive and negative consequences of carrying out the behavior, perceptions of what others think of the behavior, and perceived difficulties associated with carrying out the behavior. We integrated the data from these diverse settings, using a thematic framework analysis (Ritchie & Spencer, 1994) to first identify all the themes emerging from the whole range of contexts and then determine in which contexts each of the themes occurred. In the data collection and analysis, we focused particularly on gaining insights into reasons for nonparticipation in interventions.

Methods

Design and Procedure

Semistructured interviews lasting generally between 30 and 60 minutes were carried out with community-living older adults in Denmark, Germany, Greece, Switzerland, The Netherlands, and the United Kingdom. Prior approval for carrying out the interviews was granted by the relevant local ethical committee in each country. The interviews were conducted in participants' own languages and were audiotaped, transcribed, and then translated into English so that we could perform an integrated analysis.

To maximize the diversity of the sampled views, we interviewed people who had taken part in a variety of falls-related interventions (including people who had completed the intervention and who had not adhered to the intervention), people who had refused to take part in an intervention, and people who had not been offered a falls-related intervention. We also explicitly recruited people with a wide age range and who lived in different circumstances (e.g., in good and poor health, with and without a history of falling, from an urban and rural home, living alone or with family members, and with different levels of education).

We collaboratively developed the interview schedule and then translated it into the relevant European languages. The interview schedule was based on the theory of planned behavior (Ajzen, 1991) and covered the following issues: experiences of falls-related interventions; thoughts and feelings about interventions; beliefs about the advantages and disadvantages of interventions; factors that encourage participation in interventions or make participation difficult; and views of other people concerning the interviewee's participation in interventions. People who had previously been offered interventions were asked about their attitudes toward those interventions (see next section for a description of the range of interventions that interviewees discussed). People who had not previously been offered falls-related interventions were asked what their attitudes would be if they were offered a multifactorial intervention. This intervention was described on a card presented to the interviewee that stated the following: “These are the kinds of help that are sometimes offered to older people to reduce the possibility of falling over and being injured: training in exercises that make your legs stronger and improve your balance; medical check-up to make sure you have all the right treatments (e.g., medication, spectacles etc.); advice on how to make your lifestyle and home safer (e.g., how to do activities safely, avoid hazards in the home).”

Participants and Settings

Sixty-nine people (50 women and 19 men) aged between 68 and 97 years responded to oral and written invitations to be interviewed about their attitudes toward falls-related interventions. Table 1 shows the characteristics of participants from each setting. Just under half of the participants (30) had fallen in the past 12 months, two thirds (46) had been offered an intervention, and half (32) had taken part in an intervention.

We recruited the UK interviewees from patients who had been referred to a group-based intervention run by physiotherapists following a fall, and who were then expected to carry out exercises at home following completion of the group-based intervention. The German interviewees also all had experience of a group-based intervention run by physiotherapists, and they also were mainly referred following a fall. The Swiss participants had a range of experiences with interventions. Some had taken part in interventions including education, ergonomic advice, gymnastics classes for older people, and physiotherapy. Others had never been offered interventions. The Danish participants also had mixed experiences with interventions. Some participants had never been offered interventions, whereas other participants had been offered a group-based falls-prevention intervention run by physiotherapists to which they had been referred following a fall. Three of these participants attended this intervention and five did not. In the Netherlands, interviewees were recruited from a group of people who had been invited to take part in a cognitive behavioral group intervention to reduce fear of falling and associated activity restriction (Zijlstra, van Haastregt, van Eijk, & Kempen, 2005), but they had either declined the offer or ceased to attend after one or two sessions. The Greek participants were a community sample who had never been offered any falls-prevention interventions.

Data Analysis

We employed framework analysis (Ritchie & Spencer, 1994) because this method is suitable for systematically and comprehensively applying an analytic framework to a large quantity of qualitative data. We used Atlas.ti, version 4.1 (Muhr 1997), a qualitative data-analysis software package, to support the data analysis. Two of us developed the framework for the analysis, and we agreed on the indexing, charting, and mapping of the dataset (see the subsequent information). All of us discussed and approved the framework, findings, and interpretation of the analysis, and we agreed that the findings reflected the data we had collected from our subsamples.

There were six key stages in our analysis: familiarization, identification of the thematic framework, indexing, charting, mapping, and interpretation. Familiarization involves reading and rereading the interview transcripts to achieve immersion in the data and identify themes and ideas. We then developed a thematic framework by reviewing the notes developed during the familiarization process and identifying key themes and subthemes to create a hierarchical framework. The theory of planned behavior did not offer a reliable framework for coding the data because of the difficulty of reliably distinguishing between perceived negative aspects of interventions and perceived barriers to participating in interventions. Instead, we used four key a priori categories to structure emerging themes into factors promoting uptake and adherence, and barriers to uptake and adherence. (We did not include data from the Netherlands and Greece in the analyses of adherence as the interviewees from these settings had not had sufficient experience of participation in interventions.) We then systematically applied the framework to all the data (known as indexing), concurrently modifying and refining the framework to maximize the grounding of the framework in the data. Charting involved indicating how many interviewees had made statements relating to each theme, and in which settings each theme had been identified. This procedure allowed us to provide a broad indication of the importance of themes in the whole sample and to confirm whether themes occurred in a wide range of settings. The results of the process of indexing and charting are summarized in four tables (see Tables 2 to 5) that list all the themes and subthemes identified within each of the four main coding categories.

The final processes of mapping and interpretation involved exploring patterns and key issues by making comparisons and developing explanations that were grounded in the data. At this point, we paid particular attention to examining any qualitative differences between samples from different settings, and between participants who did or did not wish to take part in interventions. However, we had decided prior to data collection that we could not undertake a multidimensional analysis (Ritchie & Spencer, 1994), which involves comprehensively mapping individuals or attitudes onto typologies on the basis of two or more relevant factors (e.g., country, intervention type, and participation status). We made this decision because it was not possible to interview a sufficient number of participants corresponding to each combination of the many factors that might influence uptake and adherence to permit reliable estimation within subgroups of the proportion of interviewees mentioning each theme (see the introduction for a discussion of potentially relevant factors).

Results

Factors Promoting Uptake and Adherence

The themes linked to uptake of a falls-related intervention could be categorized into anticipated benefits of taking part, predisposing factors (such as personal characteristics and previous experiences), and factors that precipitated the final decision. It is evident from Table 2 that very similar themes tended to recur in all the settings sampled. The exception appears to be the United Kingdom, but this is simply because UK participants talked mainly about progressing from group-based to home-based interventions. Consequently, although the factors influencing their decisions to carry out home-based exercise were very similar to the reasons given for taking up interventions in the other settings, we coded their responses as most relevant to adherence to the transition from group-based to home-based exercise, rather than to uptake of an entirely new intervention.

The positive benefits from taking part in an intervention that motivated interviewees were wide ranging. Rather than focusing solely on reducing falls risk, participants hoped for improvements in general health, strength, and mobility that were implicitly, or sometimes explicitly, linked to maintaining independence. For example, a German woman, aged 85, gave this explanation: “I was willing to do anything that would enable me to carry on living as I was at the time—independently—being able to do my shopping, looking after myself, that was a huge motivation.” Psychosocial benefits also were frequently expected, such as learning new things, meeting people, and improving confidence, appearance, and mood.

The principal predisposing factor mentioned was personal lifetime experience of a variety of forms of exercise, including at school, in adulthood, and recent experience of other types of rehabilitation. The most important precipitating factor was a personal invitation to take part in an intervention, especially from a health professional. Other predisposing factors were perceived vulnerability or need, social approval and encouragement, and convenience. Risk of falling also was often reported as a factor precipitating uptake, although this was partly because many interviewees had been invited to take part in an intervention as a result of their high risk of falling.

The experience of a wide range of positive benefits from the intervention appeared to be an important motivation for adherence across all settings (see Table 3). Reduction in falls risk was less often mentioned than enhancement of physical capabilities and functioning and enjoyment of the activities. Several forms of social support also were widely reported as helpful. Family members could provide practical help (e.g., with transport or by supervising the exercises) and encourage participants by appreciating their achievements. Participants in groups or classes benefited from support from the group leaders and peers. A 76-year-old British woman who had fallen repeatedly in the past year gave this explanation:

You can try something because you know that there is somebody there watching, or to help you, or to help you up or whatever … they [classes] are good for seeing what other people have to cope with as well and how they can manage, you see somebody else struggling to do something else, or to do it a different way or whatever, it's quite a complex thing I think. And they [the supervisors] never look away; they never take their eyes off you.

Barriers to Uptake and Adherence

A majority of interviewees across all the settings actually denied that they needed a falls-related intervention (see Table 4), despite indications from their interviews of the presence of risk factors such as poor mobility, advanced age, and previous falls. Although low perceived risk was given as a reason for refusing participation in an intervention, some who agreed to take up an intervention also denied that they were at risk. Many stated that intervention was needed only for people at higher risk than themselves, whereas others believed that they were already taking sufficient preventive action (although few were aware of or taking all the recommended preventive measures). For example, a very physically active Danish woman, aged 82, admitted that there was a dangerous slippery rug in her home, but she still protested:

How could it [her home] be safer? I clean the windows myself and take care of the apartment. I have been offered help but I don't feel I want to have any yet. Perhaps the time may come, but while I feel I can take care of myself there are other people that need it more than me.

Some people felt that offering advice to adults who saw themselves as experienced and competent was potentially insulting or upsetting. A Greek man, aged 71, made this suggestion:

You should be very careful about the way you would approach old men and tell them that they might need to participate in this. You wouldn't like to make someone feel depressed. Not everybody accepts his age and his state. If you told a 30-year-old man “don't climb the ladder like this because you may fall,” he may accept it better than a 70-year-old man, who would say “I have climbed this ladder for 50 years, don't tell me that I will fall.”

Reluctance to be viewed as old and disabled also could negatively influence participation in groups or classes. For example, an 83-year-old Swiss woman who had had several fractures from falling nevertheless made this complaint:

It [the exercise class] was really boring. Let's say, it was—I shouldn't say this—it was too old for me! I kept telling myself “Maybe in five or six years it would be perfect” but at the time I was 80 years old, I felt that I was still too fit to be in that group …. I imagined that that kind of stuff was for people in old folks' homes.

Although some people found the classes too easy, others found them too physically or mentally challenging, often because of additional health problems such as hearing loss, incontinence, or poor memory. A Dutch woman, aged 78, who had been hospitalized following a fall and had breathing and memory problems, found the group situation problematic:

At the course I thought, “Do I really have to go through with this? I can hardly get air and can hardly talk” and that wasn't very nice for me because the others are all looking at you …. When I looked around and saw those people, it made me feel sicker.

Practical barriers to participation (e.g., time, cost, and lack of transport) were described by participants in all the settings sampled. For older people with limited physical and financial resources, traveling to attend an exercise class could require a substantial commitment of time and effort. Poor health (e.g., heart and back problems) also could pose an obstacle to an individual's uptake and adherence (see Table 5), sometimes directly preventing participation but sometimes indirectly preventing it through a person's unfounded anxiety about the consequences of exercising.

Discussion

The views expressed by older people appeared remarkably similar across the different contexts and countries sampled. This permits us to draw conclusions and make practical recommendations that may have relevance to a wide range of populations and settings.

The main reason given for nonparticipation was low perceived need, linked to a denial of risk of falling. Therefore, it might seem logical to attempt to increase the awareness of falling risk among older people in order to increase their participation. However, research into responses to falls-prevention messages suggests that the risk of falling is seen as obvious, whereas the denial of personal risk expresses a determination to maintain an active, competent, and independent lifestyle and identity (Yardley, Donovan-Hall, Francis, & Todd, 2006). Consequently, a more effective strategy may be to try to convince older people that this goal is entirely compatible with undertaking falls-prevention activities. This could be achieved by emphasizing the multiple immediate benefits of strength-and-balance training, which were widely reported as reasons for continued adherence, such as a general improvement in mobility, health, confidence, and mood, as well as interest and enjoyment. This approach would be consistent with the findings of research into older people's attitudes to general exercise, which also has revealed a dislike of health promotion messages targeted specifically toward older people, and indicates that older people are motivated to exercise principally by psychosocial rather than health benefits (Finch, 1997; Ory, Hoffman, Hawkins, Sanner, & Mockenhaupt, 2003; Stead et al., 1997). Even changing the name of falls-related interventions might be helpful; in a recent survey in Australia (Snodgrass, Rivett, & Mackenzie, 2005), older people proposed that the term falls prevention should be replaced by a more positive message such as the Better Health Club.

As well as publicizing the multiple benefits of strength-and-balance training, it may be possible to design interventions to maximize these additional benefits. Group training can provide a safe environment for enjoyable and sociable activities. However, it is clear from this study that group-based training is unsuitable for many older people, who may dislike participating in a group or may find it difficult to attend as a result of practical barriers relating to time, effort, cost, and transport. It may be helpful, therefore, to provide support for a variety of different, enjoyable interventions, both group based and home based, to match individual differences in capabilities, lifestyles, and preferences.

Advice and encouragement from health practitioners (and, to a lesser extent, family and peers) had a very strong influence on attitudes toward participation. Currently, referral to interventions is often reserved for high-risk individuals, frequently after they have already had a serious fall, and this strategy can result in high uptake rates and good outcomes. However, advice to those at high risk sometimes emphasizes the restriction of activity, and this can increase fear of falling and undermine independence (Gillespie, 2004; Ward-Griffin et al., 2004). Instead, it may be beneficial to encourage all older people to undertake strength-and-balance training, because our findings indicate that previous experience of carrying out exercise is a factor promoting participation in falls-prevention interventions, whereas very old and frail individuals may have difficulty adopting these activities for the first time. There is preliminary evidence that population-based approaches that encourage all older people to engage in falls-prevention activities (including hazard reduction and appropriate physical activity) can be effective in reducing falls-related injuries (McClure et al., 2005).

In summary, our findings suggest that, to promote uptake and adherence to an intervention, it may be helpful to involve health professionals in referral to the intervention; maximize and emphasize its potential to contribute to general health, mobility, independence, confidence, and enjoyment; and offer a choice from a range of individual and group-based activities. Although the involvement of health professionals in referrals is already common, it is not yet routine to offer a choice of individual or group-based activity, and most interventions are presented explicitly as falls-prevention programs rather than interventions with the potential to offer a much wider range of immediate benefits. However, further research is required before definitive recommendations can be made. The design of this study sought to maximize the diversity rather than the representativeness of the sample, in order to elicit a wide range of views. Now that common attitudes to interventions have been identified in this way, population surveys in representative samples are needed to establish their prevalence. Quantitative research also is needed to test the associations between attitudes, participant characteristics, and participation rates in the context of different interventions. Quantitative research would permit reliable differentiation of the effects of the different elements of the theory of planned behavior, which did not prove possible in this qualitative analysis, and it could be used to determine which factors are more relevant to uptake and which to adherence. Finally, although qualitative research can describe the reasons people give for participation or nonparticipation in interventions, controlled trials are required in order to determine whether uptake and adherence can be improved by addressing these reported barriers and facilitators.

References