Introduction

Hip fractures are common in older adults, with UK hospitals admitting 75,000 older adults with hip fractures annually [1]. Whilst most individuals recover gait, balance, and both basic and instrumental activities of daily living (ADL) after a hip fracture [2, 3], a substantial proportion do not return to pre-fracture levels of physical function or activity [3,4,5]. Similarly, reductions in quality-of-life continue to occur in the months following hip fracture [4].

People with hip fractures have defined ‘recovery’ as returning to pre-fracture activities [6] which require being mobile in various life spaces. This in turn necessitates adequate physical capabilities and confidence about one’s balance [7, 8]. However, most patients report concerns about falling in the first three months after a hip fracture [9]. Concerns about falling refer to “a lasting feeling of dread and apprehension about situations that are believed to threaten or challenge balance” [10]. These often lead to the restriction and/or avoidance of physical and social activities [10,11,12] which can then trigger a downward spiral of deconditioning, increasing physical frailty, falls, and social isolation [10, 13,14,15].

Trajectories of concerns about falling following hip fracture are complex. Evidence suggests the initial increase in concerns about falling at 4 and 8 weeks post fracture typically reduces by 3 months [9, 16, 17], although some individuals have concerns that persist for at least 6 months [18, 19] and up to 13 months after hip fracture [9]. A study investigating physical activity trajectories in the English Longitudinal Study of Ageing reported that older age and greater frailty were associated with lower physical activity participation after hip fracture [20]. To our knowledge, no review to date has investigated factors associated with activity restriction after hip fracture.

Several studies have identified factors of concern about falling after hip fracture surgery, such as pre-fracture activity, falls history, living alone, taking over four medications, post-fracture mobility and difficulties with basic ADL [18, 21]. A recent review reported that concerns about falls after hip fracture are consistently observed among people with poorer pre-injury physical function [9]. Concerns about falling after hip fracture have been associated with further falls [21], institutionalisation [21], low mood [18], lower functional abilities [18] and poorer outdoor mobility recovery [21]. Many studies exploring this topic focused on quantitative data collected at varied time points (e.g., from hospitalization up to seven years after discharge), and have not included qualitative studies.

The current review aims to synthesise the quantitative and qualitative evidence on factors of concerns about falling and activity restriction after hip fracture surgery in the community. We decided to focus on factors captured after discharge to better identify those for whom concerns about falling may be more severe and longer lasting, potentially leading to poorer outcomes. This review will enable a better understanding of who is at risk of developing concerns about falling following a hip fracture for future targeted interventions.

Methods

Design

We adopted a mixed methods systematic review design to identify factors that may be perceived by individuals in qualitative studies and/or quantified in observational studies. These studies are likely to identify different aspects of concerns about falling after hip fracture, so we followed a convergent segregated approach [22] where both quantitative and qualitative study designs are considered of equal importance and synthesised simultaneously and separately [22]. We registered the review on the International Prospective Register of Systematic Reviews (PROSPERO ID: CRD42022338881) and reported it in adherence with the updated referred Reporting Items for Systematic Reviews and Meta-Analyse (PRISMA) statement [23].

Eligibility criteria

We included observational and qualitative studies collecting data in the community, and targeting older adults (mean or median age of 50 years or older) who had previously undergone surgery for a non-pathological hip fracture. We included studies investigating the relationship between the presence/absence of any prognostic factors and ‘fear of falling’, ‘falls-related self-efficacy’, ‘activity avoidance’, ‘activity restriction’, and/or ‘balance confidence’ across care settings. We used ‘fear of falling’ as the umbrella term for our searches, as this term is most common in the literature prior to the publication of the recent World Falls Guidelines where the term ‘concerns about falling’ was instead recommended [24]. Lastly, we excluded intervention studies, studies limited to inpatient settings, conference proceedings, editorials, commentaries, case-studies, case-series, and non-English language studies.

Search strategy and screening

We searched Medline, Embase, PsycInfo, PEDro, CINAHL and the Cochrane library from database inception to the week commencing June 17th, 2022. We developed a search strategy using a combination of free text words and controlled search terms for hip fracture and fear of falling and/or activity restriction (Supplementary Appendix I). Hip fracture [25, 26] fear of falling [27], prognostic factors [28] and qualitative [29] terms were adapted from published search strategies. Title, abstract and full-text screening were completed in duplicate (ML, AC and SG) using the platform Covidence [30]. Reference lists of relevant studies and reviews were hand-searched [31]. Disagreements were resolved by consensus.

Data extraction, synthesis, and appraisal

Following a convergent segregated approach, we extracted, synthesised and integrated observational and qualitative papers simultaneously and separately [22]. Primary mixed methods studies were classified as individual observational and qualitative studies.

For observational studies, we extracted authors’ names, publication year, country, study design, eligibility criteria, participant description, sample size, analysis approach, prognostic factors (definition and timing of measurement), outcomes (definition and timing of measurement), effect estimates and measure of dispersion, and main narrative results, in duplicate (ML, AC, SG). Observational studies were heterogeneous with respect to associations explored, outcomes, and reporting of results and not amenable to meta-analysis [32]. Therefore, we synthesised observational studies narratively [33] by: (i) a preliminary synthesis focused on effects’ size and direction, and of any patterns arising; (ii) exploration of the relationships within and between reports through tabulation of studies’ results and characteristics; and (iii) assessment of the robustness of studies through quality appraisal in duplicate (SG, RMC) [33]. We followed the Quality in Prognostic Studies (QUIPS) tool to assess the risk of bias in study confounders, participation and attrition, prognostic and outcome measurements, and statistical analysis and reporting [34]. We assigned each study a ‘low’, ‘moderate’, or ‘high’ risk of bias in each of these fields.

For qualitative studies, we captured descriptive data using a standard data extraction form [35]. One author (SG) used a selective approach [36] to extract relevant findings data on factors perceived to influence concerns about falling and/or activity restriction. ‘Findings data’ was considered from text and tables published in the ‘Results’ section [35]. Subsequently, SG followed a narrative synthesis approach [37] using NVivo (version 12). To calibrate data extraction, a subset (n = 4) of studies were extracted and analysed independently by RT. On comparison, similar themes were yielded. Both reviewers followed the following steps: (i) Inductive free coding, line- by-line, of data (disregarding the research question) (ii) Grouping free codes into descriptive themes, (iii) Deriving analytical themes by inferring concerns about falling and/or activity restriction factors from descriptive themes, and (iv) assessment of the robustness of studies through quality appraisal in duplicate (SG, RT) using the Critical Appraisal Skills Programme tool (CASP) for qualitative studies [38]. With the CASP checklist, we answered 10 questions regarding each study aims, methodology, design, recruitment, collection, reflexivity, ethical issues, data analysis, findings, and implications discussed. We assigned each study a ‘Yes’, ‘No’, or ‘Uncertain’[38].

Factors identified in descriptive themes (qualitative studies) and associations (observational studies) were then classified by SG as Physical, Psychological, Environmental, Care, or Social factors (Fig. 1). We did not transform data (i.e., from qualitative to quantitative), but instead analysed and provided a synthesis of both types of evidence [22].

Fig. 1
figure 1

Extraction, synthesis and appraisal of qualitative and observational studies following a convergent segregated approach

Full size image

Results

Search

We identified 485 papers following de-duplication and 10 from hand searches of reference lists (Fig. 2). In total 375 were excluded on title and abstract screening, and a further 101 were excluded following full-text screening. We included 19 papers in the review, representing 1 mixed method [39], 9 qualitative [8, 40,41,42,43,44,45,46,47] and 9 observational [19, 48,49,50,51,52,53,54,55,56] studies.

Fig. 2
figure 2

PRISMA flow diagram

Full size image

Study characteristics

Observational

We included nine observational studies and one mixed-method study of cross-sectional [39, 52,53,54], prospective [19, 48,49,50, 55] and retrospective [51] designs, conducted in Europe [39, 48, 51, 53, 54], North America [49], Japan [50, 52] and Australia [19, 55] (Supplementary Appendix II). The studies reflected analyses from 1330 older adults after hip fracture (sample size range 33 [39] to 263 [49]). Participants were mostly women (68% [51] to 100% [52]), and the mean age ranged from 64 [52] to 83 [48] years old.

Qualitative

We included nine qualitative studies and one mixed-method study which were conducted in Europe [39,40,41,42,43,44, 46, 47] Brazil [45] and Australia [8] (Supplementary Appendix II). The research was of inductive [41] phenomenological [8, 45,46,47], grounded theory [8, 43, 44], or unspecified [39, 40, 42, 46] design. The studies reflected 150 older adults after hip fracture (sample size range from 4 [39] to 31 [40]). Most participants were women (from 64.5 [42] to 88% [41]), and the mean age ranged from 74 [46] to 85 [40] years. One study interviewed both carers and patients [42]. Four interviewed participants multiple times [40,41,42, 46]. Two publications [43, 44] analysed the same dataset.

Outcome

Observational

Outcomes were measured with the Activities-specific Balance Confidence scale (ABC) [51, 53, 54, 57], the Falls Efficacy Scale (FES) questionnaire [50, 58], the FES-International (FES-I) [48, 59], the short FES-I [19, 49], a modified FES [52], both the ABC and FES [55], or both the FES-I and the modified Survey of Activities and Fear of Falling in the Elderly [39]. Outcomes were assessed at hospital admission [48, 49], 2 to 6 weeks post-fracture [19, 49, 50, 53], 3 to 6 months post-fracture [19, 39, 48, 50, 53, 55], and up to 7.5 years post-fracture [51, 53, 54]. One study did not specify when assessments in the community took place [52].

Qualitative

Studies explored concerns about falling and/or activity restriction through a variety of research questions, such as the overall impact of a hip fracture [40,41,42, 47]; or how mobility [8], adaptations [41], concerns [43, 44], sedentary behaviours [45, 46], and fear of falling [39] changed post-fracture. Data was collected during hospital admission [8, 40], 2 to 12 weeks post-fracture [8, 40, 42,43,44, 46], 3 to 6 months post-fracture [39,40,41,42, 46], and over 6 to 24 months post-fracture [40, 45, 47].

Critical appraisal

Observational

Six studies [39, 48,49,50, 52, 55] had a high risk of bias in at least one of the ten QUIPS domains. Four of these [39, 48, 50, 55] were on confounders and two were on study attrition [49, 55]. Moderate risk of bias was determined for all studies in at least one domain. Most (n = 9) [39, 48,49,50,51,52,53, 55, 56] were on the prognostic factor measurement, five [49, 52,53,54, 56] on study confounding, and four [39, 48, 50, 55] on statistical analysis and reporting.

Qualitative

Almost all studies clearly stated research aims that were adequately answered by qualitative studies, appropriately recruited participants, and collected data (CASP item 1, 2, 4, 5). Most provided detailed information on the analysis steps, provided representative quotes, and discussed findings contributions and usability/future steps (item 8, 9, 10). Over half of the studies (n = 6) failed to critically examine researchers’ relationships with participants (item 6) [8, 40, 41, 43,44,45]. Four failed to justify their research design (item 3) [8, 40, 41, 44] (Supplementary Appendix III).

Factors

Tables 1 and 2 show study findings related to concerns about falling and activity restriction. Table 3 presents a summary of all factors classified as physical, psychological, environmental, social or care factors. Evidence from both study designs is summarised below.

Table 1 Findings of observational studies
Full size table
Table 2 Findings of qualitative studies
Full size table
Table 3 Physical, psychological, care, environmental, and social factors identified
Full size table

Factors of concerns about falling and activity restriction

Observational studies reported more concerns about falls and activity restriction among participants who fell after their surgery, lived with comorbidities or had poorer mobility and functionality post-fracture. In qualitative studies, participants attributed low mobility pre-fracture, having fallen again after surgery, fatigue and lower strength, as influencing factors for concerns and reduced activity (Physical). One primary source summarised these concerns:

“Four of the interviewees had experienced a new fall after discharge, which they thought had added to their fear of falling. The deteriorated ability to move and walk had made them lose power and strength, mentally as well as physically, and they experienced that they had become more tired than before.”[47]

In observational studies, concerns about falls were seen among participants with higher scores on anxiety and neuroticism scales. Authors of qualitative studies noted that concerns about falling and activity restriction co-existed with diminished confidence in recovery and own abilities to engage in physical therapy and overcome challenges. Participants who acknowledged concerns and associated restricted activity reported being hypervigilant of everyday activities that could lead to a fall, negative connotation of older age, the fracture event, pre-fracture abilities and the recovery journey (Psychological).

“Some interviewees described the shock after the injury in words such as ‘everything became a mess’, and that the injury had brought feelings of insecurity and distress. ‘You think, why didn’t I put the light on when I got up [at night]? So, I’m very careful now, almost excessively so.’”[47]

Across qualitative studies, participants worried about falls and restricted activities outdoors because of poor weather, uneven surfaces, the need to plan in anticipation, and the lack of accessibility in public places and transport. Indoors, participants described higher concerns about falling and activity restriction when activities required climbing stairs or mobilising in the dark. Basic activities of daily living such as toileting were recounted to increase concerns, especially early post discharge (Environment).

“One year later most of the patients had regained their previous level of physical functioning and were able to perform almost all usual activities of everyday life except for some outdoor activities such as taking the bus or biking because they were ‘afraid of falling’.”[40]

Participants in qualitative studies who expressed more concerns about falling and activity restriction often suggested they did not fully understand or recall information provided. Some felt that healthcare professionals’ suggested activities might lead to a new fall or that they did not fully understand their needs (Care).

“Anna was dependent on help in her home and felt overlooked when busy staff did not recognize her needs: ‘and then all of a sudden they forget a lot of things for example when I used a walker, they could forget to put it where I could reach it and there are a lot of things like that. That’s no good’.”[46]

Qualitative studies also noted concerns about falling and activity restriction among people living alone, with less involvement in social activities (due to both their own volition and others withdrawal), or with relatives who restricted activities deemed unsafe. This impeded individuals’ ability to challenge themselves and rehabilitate at their own pace[39, 44] (Social).

“The informants had become more bound to their homes and dependent on others to go outside. They saw friends and relatives less often than before and it was up to their relatives to take the initiative.”[39]

Factors not associated with concerns about falling and activity restriction in observational studies included gender, cognition, positive and negative affect scales, the use of multiple medications and the type of surgery/fracture. Factors with conflicting evidence for an association included pain (n = 3 association [50, 52, 54], n = 2 no association [48, 49]), age (n = 2 association [48, 53], n = 2 no association[49, 52]), depression scales (n = 1 association [48] n = 2 no association [49, 52]), and time since fracture (n = 1 association [19], n = 2 no association [52, 53]). In qualitative studies, balance and pain have conflicting evidence—with some participants reporting it contributed to activity restriction and others saying it did not (Table 3).

Factors to overcome concerns about falling and activity restriction

In observational studies, participants with better balance, strength, and mobility showed less concerns about falling (Physical). In qualitative studies, participants who discussed self-determination to remain active and acknowledged that this may ‘look different’ compared to their pre-fracture activity [39, 41, 45,46,47] – with respect to pacing [39, 41, 45,46,47] and behavioural adaptations [46] (e.g., only walking over smooth surfaces and for shorter distances)—reported less activity restriction. Many of these participants also celebrated previous challenges overcame or current positive aspects of life [47] (Psychological).

“Being persistent, creative, positive, vigilant, and thoughtful supported a sense of being, and experiences of progress maintained hope and self-confidence. Responsibility was a matter of finding solutions to problems and having duties. Karen had managed baking cookies for my second visit: ‘I tell myself, you HAVE to try, and then when a full baking sheet is ready, I go and sit down for a while. Then back to the oven again. It can take a long time, but I have nothing but time.’ [46]”

Qualitative studies pointed out that walking aids and adaptations to the home helped to increase activity and reduce concerns, but required practice and some were considered problematic. Good weather, adequate public transport services and being able to drive, were considered key to increase activity (Environment).

“According to the informants, a walker reduced fear of falling and was necessary at least when walking outside. It was a source of security, but also a nuisance and barrier to moving freely. Another complaint was the inability to use public transportation because the walker was unmanageable in a bus or a car.”[39]

Participants who reported receiving clear advice on how to safely mobilise and engage with activities without falling, as well as on the consequences of restricted mobility, described less activity avoidance and concerns about future falls. These experiences were augmented through the regular presence of healthcare professionals [45, 46] providing positive feedback and encouragement [44, 46], acknowledging positive progress [44], and listening and acting on participant’s concerns e.g., medications review [47] (Care).

“Balancing risk safely was a consequence of being provided with adequate information. Being informed related to receiving information, feedback, advice or reassurance from healthcare professionals regarding progress. Older people had to rebuild their damaged confidence, unsure of their abilities and needed encouragement to increase their self-efficacy.”[44]

Qualitative studies described that the presence and company of others provided a sense of connection, motivation, and safety, relieved feelings of loneliness and low mood, which in turn allowed people to move in and outside the house more [45,46,47]. Others encouragement and awareness of their needed accommodations helped to relieve concerns and activity restriction too. Practically, informal caregivers helped by taking individuals to appointments, shopping, for a walk, or helped to use transport and travel long distances. In a cross-sectional study, however, greater social support was not related to concerns about falling (Social).

“For the interviewees, having a support network, whether formal or informal, was indicated as a way of overcoming dependency to move around and, especially, to walk safely, resulting in better mobility and more active behaviour. Agatha [said]: Yes, just the fact that I had someone, here with me in the afternoon, meant I already felt better. Having human warmth, having company around, I could do more things outside [...] yes, if I had company I’d go by public transport. For example, I’d go to some park, get some fresh air in the park.” [45]

Discussion

Main findings

This mixed-methods systematic review focused on factors related to concerns about falling and activity restriction after hip fracture. We report eight physical, five psychological, five environmental, three social and two care factors contributing to concerns about falling and associated activity restriction after hip fracture. The factors investigated by observational studies were weighted towards the physical, while qualitative studies identified more aspects related to the environment, care, social and psychological. Most factors were reported on by a small number of studies of varying quality.

Findings suggest that concerns about falling and associated activity restriction are more likely to be observed among patients with greater comorbidities; poorer physical and functional abilities post-fracture; less social support; accessibility issues (e.g., living out of area); a lack of, or inability to access, home adaptations; less psychological resources, and/or; poorer perceptions and experiences of the rehabilitation provided at hospital and/or home. Similarly, findings suggest that less concerns and activity restrictions post hip fracture are observed among people with better physical function; higher psychological resilience and positive affect; greater social support; adequate accessibility indoors and outdoors, and; better perceptions and experiences from formal care at the hospital and at home.

Interpretation

Findings suggest that rehabilitation designed to target physical factors (e.g., strength, function) might help with concerns about falling and activity restriction post-hip fracture. Further reductions in concerns may occur through directly addressing the psychological consequences of a fracture and acknowledging the fact that people will progress at different rates [3], especially in the first three months post-fracture when concerns about falling tend to be high (and potentially reflect an adaptive process) [16, 17]. Our findings suggest that promoting a positive mindset, in addition to building self-confidence and motivation to engage in rehabilitation, may have positive long-term effects on concerns about falling and activity restriction. Indeed, patients have indicated that support and coaching facilitate recovery in daily living after hip fracture [60]. A previous study reported increased physical activity and walking, and reduced concerns about falling, by promoting confidence and motivation for change through motivational interviewing among community-dwelling older adults who had a hip fracture [61].

Few studies examined care factors, despite these being potentially the most amenable to interventions. Where assessed, most focused on communication between healthcare professionals and patients. Effective communication was believed to mitigate concerns about falling and activity restriction by attending medical concerns, increasing access to formal care, providing positive feedback, advice on health consequences, and strategies for safe mobility and ADL engagement [43,44,45,46,47]. Effective communication strategies have been considered crucial by clinicians for engaging patients and improving outcomes after hip fracture [62, 63]. One cross-sectional study of low quality suggested no association between the use of multiple medications and concerns about falling, even though this may reflect frailty and was associated with concerns about falling after hip fracture in a recent study [18]. Further research on potentially modifiable care factors related to concerns about falls and activity restriction after hip fracture is warranted.

The persistence of environmental factors increasing concerns about falling and activity restriction in later stages of recovery aligns with the vast complexities of mobilising as an older adult [64], and the impact of environmental barriers on older adults’ activity and function outdoors [65]. The accessibility of public spaces and services such as transport does not tend to meet the growing demand of people with limited mobility and walking aids [66, 67], especially when additional safety measures are required to avoid a fall [65, 68]. Services that help people go outdoors seem invaluable for those with limited networks to reduce concerns about falling and activity restriction, particularly later post-discharge. However, rehabilitations incorporating outdoor mobility components did not show improvements in falls-related self-efficacy, possibly in part due to an absence of targeting environmental barriers related to mobility [69]. In support, an intervention providing walking maps for the local community environment improved time spent walking outdoors [70].

The current review noted support and company from family, friends, and formal care was associated with a reduction in concerns and increased activity, while relatives' fear and restrictions limited mobility and activity. Findings align with previous reports of patients, informal carers, and clinicians, stating the importance of educating and involving relatives in rehabilitation to improve outcomes across the care continuum [6, 8, 62, 71]. Nevertheless, informal carers report feeling excluded from rehabilitation, struggle to make sense of the information shared, and their relatives needs post-fracture [71]. Findings also emphasise the need and benefit of providing individuals with reduced social networks and who may withdraw from social activities, with alternative means of engagement. Interventions including carers to set goals [72] and to support discharge and home care [73], have showed reductions in concerns about falling at one [73], four [72] and 12 months [74] follow up.

Future research

Previous evidence suggests concerns about falling at three months was associated with poorer recovery outcomes, an observation that was not seen in the first four weeks [17, 75] or 12 months post-fracture [18]. This suggests that there is a key ‘window’ (between four weeks and three months) to address the fear of falling post-hip fracture, when it appears to reflect a maladaptive process. Further, effective interventions reducing falls concerns have mostly worked for individuals with higher functional ability [72, 73] and were delivered in hospital and community settings [21]. Future interventions should seek to target the factors identified by the current review that predispose an individual to concerns about falling post-fracture. These interventions may employ risk stratification for immutable factors such as pre-fracture function, comorbidities, or stairs at home. Alternatively, they may directly target modifiable factors such as low confidence, social support, or post-fracture function.

Strengths and limitations

A strength of the review was the use of both observational and qualitative evidence. This provided depth and variety to our findings, expanding on the multiple individual and external factors that impact concerns about falling and activity restriction. We captured studies on varied populations, with qualitative studies tending to focus on under-researched or more vulnerable populations. A main limitation is that we did not draw causal, prospective associations. All studies had a high or moderate risk of bias in study attrition, or analysed data cross sectionally. This issue was also highlighted in a previous review [21]. The quality of the evidence was poor limiting interpretation. For example, no association may reflect a lack of power, a positive/inverse association may occur where authors failed to account for confounding and/or a failure to report on reflexivity. Lastly, there may be different factors for activity restriction and concerns about falling, but we could not clearly differentiate between the two as studies did not report separate results. Our search criteria yielded fewer results than expected so we also hand-searched reference lists [31].

Conclusion

We observed concerns about falling and activity restriction among individuals following hip fracture with a history of falls, comorbidities, low energy, balance and functionality, who reported low confidence in their own abilities and in rehabilitation. On the contrary, participants with less concerns and activity restrictions had better strength, mobility, social support, formal care experiences, and the ability and confidence to take control over recovery (e.g., adapting behaviours, asking for help). Further, practical social support from informal and formal networks, and the accessibility of indoor and outdoor spaces, seemed essential to overcome fears and increase activity. Findings highlight patient populations who may be at increased risk of longer-lasting concerns and activity restriction resulting in poorer outcomes (e.g., low social support) or who may need more help to overcome worries (e.g., people with anxiety and other comorbidities). Findings also point to an array of potential targets to encourage activity after hip fracture such as self-confidence, strategies for safe mobility and social support from formal and informal networks.