Interventions with proven efficacy will be effective in practice only if patients adhere to them,¹ but clinical guidelines rarely make evidence-based recommendations about how best to involve patients in their implementation. This paper presents recommendations for promoting the engagement of older people in the prevention of falls.

Falls are the most common cause of accidental injury in older people, and frequently result in disability and handicap, emotional distress, and increased use of health and social services.^2,3 Clinical trials have shown that falls-prevention programmes can reduce the incidence of falls.^2–5 However, the effectiveness of these programmes is potentially compromised by the reluctance of many older people to take part, which can lead to low uptake rates and significant levels of dropout and non-adherence.⁶ The Prevention of Falls Network Europe (ProFaNE; http://www.profane.eu.org), therefore, undertook developing guidelines for maximising the acceptability of falls-prevention interventions among older people as its key objective, in order to increase their participation levels. However, the evidence base for these guidelines has proved to be weak. There is little research on adherence in the field of geriatric rehabilitation.⁷ Published falls-prevention interventions have used a variety of methods to promote adherence, but their effectiveness is difficult to determine, because they have not been tested using a randomised controlled design. Comparisons between intervention studies are problematic, because variations in the methods used to promote adherence in different studies are confounded by differences in the sample, setting and intervention, which can also affect adherence. Moreover, descriptions of interventions do not usually provide sufficient detail to permit evaluation of how their various components may have affected adherence.

A new method for developing guidelines has recently been proposed,⁸ which combines the Delphi survey method of eliciting anonymous opinions from people who are geographically dispersed with the opportunity for debate provided by the nominal group technique. This new approach was explicitly intended for use in situations where the evidence base is weak, but there is a need to make recommendations for improvements in clinical practice and research. In this situation, the evidence base must be supplemented by a consensus based on the subjective views of those working in the field, making this process explicit and transparent by reporting aspects of the consultation process such as the degree of consensus and reasons for disagreement. We utilised this approach for the development of our recommendations, modifying the method to fit the requirements of our particular context. The aim of this paper is to describe the process and outcomes of developing recommendations for promoting the engagement of older people in falls prevention.

METHODS

Generation of recommendations

The development of the recommendations commenced with a meeting of nine members and associate members of ProFaNE Workpackage 4 (a workpackage of researchers and clinicians addressing the psychological aspects of falls prevention) in September 2005. This group (which included all the authors) identified seven topics that seemed relevant to engaging older people in falls-prevention programmes.

To provide an evidence base to inform our recommendations, we first carried out two large qualitative studies and a survey of older people’s views on falls prevention.^6,9,10 These studies elicited the views of a wide range of older people in different circumstances, and examined which factors were most closely associated with intentions to undertake strength and balance training to prevent falls. We were also able to draw on a recently published narrative systematic review of older people’s views of falls prevention.¹¹ Because of the limited evidence base relating specifically to falls-prevention interventions, we decided to also draw on a wider literature, including basic psychosocial theory and research into participation in other preventive healthcare activities.

One or more individuals took responsibility for developing a draft recommendation relevant to each topic, and for compiling an evidence base for each recommendation. These were then developed and revised in discussion with the whole group.

Consultation process

This process involved the further development and refinement of the recommendations via iterative discussions involving the core group (the co-authors), and a series of wider consultations both within and outside the membership of ProFaNE. Table 1 summarises the process, with further details provided below. A description of the key changes to the recommendations made during this process is given in the Results section.

1. The initial meeting, during which seven key topics were identified and the core group established, formed part of a larger meeting of all workpackages in the ProFaNE network. The key topics were therefore presented to the 49 people working in the field of falls prevention who attended the ProFaNE meeting, to obtain rapid feedback regarding the proposed topics.

2. Seven outline recommendations were subsequently developed and posted on the Workpackage 4 discussion board on the ProFaNE website, for refinement through discussion with all members of Workpackage 4.

3. Once the draft outline recommendations had been agreed within Workpackage 4, we carried out a formal website consultation, to permit anonymous rating and commentary on our proposed recommendations from a wide range of professionals working in falls prevention (including medicine, nursing, physiotherapy, occupational therapy, psychology, management and so on). For 6 weeks, the recommendations were presented on the ProFaNE website to 171 researchers and clinicians working in the field of falls prevention, whose details were registered on the ProFaNE website. Any of these ProFaNE members logging onto the website during this period were automatically invited to indicate their level of agreement with each recommendation on a 9-point scale⁸ ranging from 1 (strongly agree) to 9 (strongly disagree), and to give a brief written explanation of the reasons for each of their ratings. Of the 35 respondents, 17 were clinicians, 13 were academics and 5 belonged to other occupations (eg, epidemiologist, hospital manager, PhD student). Most people (n = 19) had between 2 and 5 years of experience working in the field of falls prevention, 11 had >5 years experience and 5 had <2 years experience.

4. We then carried out a face-to-face consultation with a large group of people working in the field of falls prevention, five of whom had also participated in the internet consultation process, in order to permit in-depth debate about the recommendations. A half-day workshop was convened during an annual meeting attended by 31 people, comprising 12 clinicians, 12 academics and 7 people who listed other occupations (eg, combined clinician and researcher). Most (n = 19) had >5 years experience in the field of falls prevention, nine participants had 2–5 years of experience and two had <2 years of experience. The core group member with primary responsibility for each recommendation gave a 10 min presentation outlining the recommendation, the evidence on which it was based, and the ratings and comments from the website consultation. Extensive discussion followed each presentation. To ensure accurate consideration and reporting of the comments made at the workshop, it was audiotaped and transcribed, and notes taken to aid interpretation of the transcript.

5. Written drafts of the recommendations and the evidence on which they were based were then completed by the co-authors and circulated within this core group. The co-authors critically discussed the content and presentation of the recommendations and agreed formats for dissemination in a final meeting. After a final draft of each recommendation had been circulated, LY took responsibility for collating the recommendations into draft documents for dissemination, incorporating comments on the drafts from all co-authors.

RESULTS

Tables 2 and 3 present the final recommendations, a summary of the evidence on which they were based, and the website ratings of the level of agreement with each recommendation. Agreement with the early draft of the recommendations presented on the website was already uniformly high, with a median agreement score at or just one point lower than the maximum agreement. Although at least 75% of the ratings for every draft recommendation were firmly in agreement (scoring at least 3 out of 9), a few individuals registered mild disagreement with some recommendations (giving isolated scores of 6 or 7 out of 9). Constructive critical comments made during the website consultation and at other stages of the consultation process resulted in further refinement of the recommendations. The important changes are outlined below.

Modifications of the recommendations for promoting the uptake of falls-related interventions

The first recommendation provoked discussion about the effectiveness and cost effectiveness of falls-related interventions for the general population of older people. Since the evidence base for this is not yet conclusive,^2,4,5,12 the final recommendation was cautiously worded to suggest that awareness at a population level regarding the potential benefits of these activities should be raised.

The earliest draft of the second recommendation suggested emphasising the positive (physical) benefits of falls prevention rather than a reduction in undesirable events (ie, falls). Discussion in a wider group the following day immediately revealed that this formulation had the potential to be misinterpreted as suggesting that a reduction in the risk of falls should not be mentioned at all. Similarly, a draft recommendation suggesting that the multiple benefits of physical activity should be promoted proved ambiguous (some people took it to refer to the benefits of general physical activity rather than the general physical benefits of falls-prevention exercises), and was therefore merged with the second recommendation.

Modifications of the recommendations for promoting the uptake and adherence to falls-related interventions

Originally, the third recommendation proposed that all forms of social encouragement should be maximised, but comments and debate highlighted the risk that excessive social pressure could be insensitive to the autonomy of the individual, and so a less strong formulation was adopted. A further comment was that it would be useful to prioritise the most effective forms of social encouragement. However, we decided that we could not do this because there is currently no directly relevant research on which to base such a prioritisation, and the most effective form of encouragement may vary depending on the context and priorities of the individual (ie, the fourth recommendation).

The wording of the fifth recommendation was changed to use the term “self-management” rather than the term “self-help”, and to note that the health professional must supervise activities to ensure safety. This was because commentators noted that not all older people were able and willing to undertake falls-prevention activities independently, and that the health professional played a crucial role in ensuring the right level of support. Comments on the sixth recommendation focused on the weakness of the evidence on which to base recommendations for using specific methods to promote adherence in falls-related interventions, and so the wording of this recommendation highlights the need to assess the processes that influence adherence in this context.

DISCUSSION

The recommendations we developed highlighted the need for: disseminating the message that physical activity can improve balance; promoting the immediate benefits of interventions; ensuring that interventions support a positive identity and confidence in self-management; tailoring interventions to the specific situation and values of the individual; using social encouragement to engage older people; and employing validated methods to maintain longer-term adherence.

A number of limitations of the process of developing the recommendations need to be taken into account. A positive feature of the process was that the initial development of the recommendations was informed by substantial qualitative and quantitative studies of the views of a wide range of older people.^6,9,10 However, further consultation with older people is required to confirm the acceptability and appropriateness of the final recommendations. A large number of clinicians and researchers contributed to the refinement of the recommendations, and expressed support for them. However, only a minority of those registered on the falls-prevention website rated the recommendations. Since all the experts attending the subsequent open meeting also endorsed the recommendations, we do not think it likely that the views of the non-respondents to our website consultation would have differed greatly; probably, they simply did not access the site during this period. However, only a qualitative process of achieving consensus for further refinements of the recommendations was used. Therefore, further quantitative evaluation of the views of a purposive sample of professionals working in the field of falls prevention could confirm their appropriateness.

These recommendations represent a consensus based on current knowledge and evidence, but the evidence base from which these recommendations were developed was limited, and not always specific to falls prevention. To increase the effectiveness of falls-prevention interventions, further research is needed to experimentally confirm the features of falls-prevention programmes that will encourage older people’s engagement in them.