Media: British Journal of Nursing (BJN) {Main} Edition: Date: Thursday 12, November 2020 Page: 58,59,60,6 Becoming fall-safe: a framework for reducing inpatient falls Mark Renshaw, Paula Tucker and Karen Norman ABSTRACT This article d e s c r i b e s a 10-year p r o g r a m m e of work t h a t has reduced inpatient falls rate by 4 6 % a n d how t h i s i m p r o v e m e n t has been s u s t a i n e d . The m e t h o d o l o g y applied in t h i s initiative has forced one Trust t o challenge e x p e c t a t i o n s a b o u t t h e inevitability of patient falls in hospital. This initiative has r e s u l t e d in approximately 5 6 8 fewer falls each year. Based on c o s t i n g s f r o m NHS I m p r o v e m e n t , t h e e s t i m a t e d 5 1 0 8 fewer falls b e t w e e n 2 0 1 1 a n d 2 0 1 9 have saved t h e Trust £ 1 3 . 3 million. Key w o r d s : Patient falls • Patient safety • After action reviews • Complexity s c i e n c e • Reflective practice I npatient falls are the most c o m m o n l y reported patient safety incidents in the N H S . T h e most recent Organisation Patient Safety Incident R e p o r t s ( O P S I R ) w o r k b o o k s based on data submitted to the National R e p o r t i n g and L e a r n i n g System ( N R L S ) c o v e r i n g t h e p e r i o d April 2018 to M a r c h 2019 r e p o r t e d 228 5 0 3 patient accidents in acute (non-specialist) hospitals in England and Wales.The vast majority of these patient accidents would be patient falls and the n u m b e r suggest more than 600 patients fall every day ( N H S Improvement,2020a; 2020b).With statistics like these,it is easy to see w h y inpatient falls are often viewed as ubiquitous and inevitable in hospitals. All falls, even those that do not result in injury, can cause patients and their families to feel anxious and distressed. For those w h o are frail, m i n o r injuries f r o m a fall can affect physical function, resulting in reduced mobility and u n d e r m i n i n g patient confidence and independence. Based on data presented in an N H S Improvement report on the incidence and costs of patient falls in hospitals (2017), the authors calculated that there are 5310 serious injuries per a n n u m in hospitals in England, including hip fractures and head Mark Renshaw, Head of Quality Improvement, Brighton and Sussex University Hospital NHS Trust, mark.renshaw@nhs.net Paula Tucker, Deputy Chief Nurse, Innovation and Improvement, Surrey and Sussex Healthcare NHS Trust Karen Norman, Visiting Professor, Complexity and Management Group, University of Hertfordshire, Visiting Professor, School of Nursing, Kingston University and St George's, University of London, and Non-Executive Director, Queen Victoria Hospital NHS Foundation Trust, East Grinstead Accepted for publication: June 2020 injuries, w h i c h can result in patient death. Falls in hospitals are also expensive as they increase the length of stay and may require increased care costs on discharge. N H S Improvement (2017), calculated that the average cost of a fall in hospital is ^ 2 6 0 0 . This p r o j e c t started in the financial year 2009—2010 at Brighton and Sussex University Hospital (BSUH) N H S Trust, at w h i c h t i m e the Trust's incident data recorded 1400 falls and reported a rate of 6.23 falls per 1000 bed-stay days. This rate was slightly lower than the national falls rate of 6.63 per 1000 bed days, as reported by the Royal College of Physicians ( R C P ) in 2015. T h e r e was a c o m m o n l y held assumption that falls were an unfortunate side effect of a patient's rehabilitation or an unavoidable part of their deterioration into frailty. Such cultural norms and expectations served to reinforce staff perceptions that falls were inevitable, which can then become a self-fulfilling prophecy. Background and literature review T h e j o u r n e y to reduce falls at B S U H began in April 2 0 0 9 . T h e catalyst for this w o r k was, unfortunately, far from u n i q u e : a catastrophic fall leading to the death of a patient, a bereft family, guilt-ridden staff and damning media coverage.The Trust's Chief of Safety made a c o m m i t m e n t to the family of the deceased patient to reduce the n u m b e r of inpatient falls in the Trust, mindful of the fact that tackling the problem of inpatient falls is a huge challenge. T h e falls project team reviewed C a m e r o n et al's (2010) C o c h r a n e review of interventions for preventing falls, which concluded that there are no 'quick fix' interventions that, w h e n implemented o n their own, significantly reduce the rate of inpatient falls. T h e timeline of the project is shown in Table 1. Methodology: a traditional approach In the first falls project in 2009, the falls project team applied a traditional change methodology, organising a Trust-wide launch of a falls programme, ensuring that every ward manager could attend.To emphasise the importance of this initiative, the team invited a national expert on falls to educate staff, a renowned speaker on leadership to inspire staff, a doctor passionate about reducing patient sedation as a strategy for reducing patient falls and physiotherapists and occupational therapists keen to work collaboratively with nurses. At the launch, the ward managers and occupational therapists were given individualised reports detailing the n u m b e r , rate, time, and o t h e r specific details of falls o n their wards. Similar COPYRIGHT: This cutting is reproduced by Gorkana under licence from the NLA, CLA or other copyright owner. No further copying (including the printing of digital cuttings), digital reproduction or forwarding is permitted except under license from the NLA, www.nla.co.uk (for newspapers) CLA, www.cla.co.uk (for books and magazines) or other copyright body. Article Page 1 of 8 A23797 - 3 Media: British Journal of Nursing (BJN) {Main} Edition: Date: Thursday 12, November 2020 Page: 58,59,60,6 wards were grouped together in facilitated workshops to enable variations in p e r f o r m a n c e to be compared and discussed. At the e n d of the w o r k s h o p , each ward manager left w i t h an individualised action plan, detailing h o w they were g o i n g to reduce the falls rate on their ward. Collectively, the ward managers identified 150 actions that they and their teams were g o i n g to take to reduce the n u m b e r of falls in the Trust. T h e initial results appeared promising, indicating a reduction in the m o n t h l y rate of falls. However, by the third m o n t h the rate of falls had fallen back into a pattern of normal variation. At the 3 - m o n t h follow-up meeting, the wards reviewed their c o m p l i a n c e w i t h the action plans and generated n e w ones. This p a t t e r n c o n t i n u e d for the rest of the year and, despite t h e c o n s i d e r a b l e e f f o r t e x e r t e d i n t o r e d u c i n g t h e rate o f falls, it r e m a i n e d stubbornly fixed, seemingly i m p e r m e a b l e to change (see Figure 1). T h e review of p e r f o r m a n c e in the first year divided o p i n i o n . S o m e ward managers genuinely believed progress was b e i n g made and p o i n t e d to the analogy of stars in the m i d n i g h t sky: that the harder o n e looks, the m o r e you will find, arguing that by focusing the 'telescope' on falls, m o r e r e p o r t i n g was e n c o u r a g e d . O t h e r s suggested that the evidence base for falls reduction was weak, c o n f i r m i n g their original c o n t e n t i o n that falls were inevitable. A n o t h e r g r o u p suggested that the w r o n g actions had b e e n chosen, or that staff were just n o t i m p l e m e n t i n g t h e m , or had n o t u n d e r s t o o d t h e m . With hindsight, the falls project team can n o w see that they followed a c o m m o n approach to safety and quality initiatives w h e n quality targets are not met.This was, namely, identifying best practice (guidelines, policies, care bundles, action plans etc) and m o n i t o r i n g performance to check this was done. If results failed to improve, m o r e of the same was d o n e ; that is, m o r e detailed guidelines were produced, there was m o r e frequent scrutiny, and even more action plans. Although the team believed the latter can be necessary and useful, a point raised by o n e m e m b e r made the team start to question this approach.They said: 'We've d o n e e v e r y t h i n g a g r e e d . If a d r u g didn't s e e m t o be w o r k i n g , w o u l d w e k e e p g i v i n g it to the patient?' This challenge provoked the project lead to start to question the u n d e r p i n n i n g theoretical assumptions behind this approach to quality improvement. O n e of the observations during the course of this initiative was that m u c h of the evidence base tells you 'what to d o ' . T h e real challenge is making this 'what you do' to incorporate best practice into your daily activity. By early 2010, it was clear that despite the project team's best efforts, this approach was not going to deliver a significant reduction in falls. Simultaneously, the team was beginning to consider alternative explanations for the results from the first year and developing a h u n c h that proved to be the key to u n l o c k i n g the eventual approach to r e d u c i n g the falls rate. This h u n c h went beyond locating responsibility for reducing falls primarily with ward leaders and was i n f o r m e d by insights f r o m i m p r o v e m e n t m e t h o d s inspired by complexity t h e o r y o o o •H Initial falls rate (July 2 0 0 8 — June 2009) Falls rate at end of year 1 (July 2 0 0 8 — June 2009) Initial falls rate 2nd year, 8 wards pilot (April 2 0 0 9 - March 2010) Falls rate at end of 2nd year, 8 wards pilot (April 2 0 1 0 — March 2011) Figure 1. Results of first-year Trust-wide project and second-year eight-ward pilot Table 1. Project timeline Date Event April 2009 Catastrophic fall on acute medical ward July 2009 First Trust-wide falls project launched (falls rate reduced from 6.23 to 6.07 per 1000 bed days) April 2010 Second falls project launched (eight-ward pilot—falls rate reduces from 9 . 1 to 6.6 per 1000 bed days) April 2 0 1 1 Third Trust-wide falls project launched 2 0 1 2 - 2 0 1 3 Falls rate drops below 5.5 per 1000 bed days 2 0 1 3 - 2 0 1 4 Falls rate for previous 12 months drops to below 4.2 2 0 1 4 - 2 0 1 6 Falls rate for previous 12 months drops to below 3.5 2016 Senior nurse joins a neighbouring trust 2 0 1 6 - 2 0 1 9 Annual falls rate drops below 3.4 (Suchman, 2011). T h e project team became interested in whether the behaviour of nurses on the ward could be influenced so that they were 'falls aware'.The team also speculated that, if one individual m e m b e r of staff behaved in a falls-aware manner, this behaviour might spread to other individuals w h o came into contact with this individual—in m u c h the same way as a virus spreads through a group. Buchanan (2007) highlighted that we are, by our very nature, born imitators: 'We d o n o t think entirely o n our o w n , w h a t w e believe and w h y d e p e n d s s t r o n g l y o n o u r i n t e r a c t i o n w i t h o t h e r s . . . [ T h e ] r o o t s o f i m i t a t i v e b e h a v i o u r . . . s e e m t o b e a u t o m a t i c and u n c o n s c i o u s , and instinctual, h a r d w i r e d i n t o COPYRIGHT: This cutting is reproduced by Gorkana under licence from the NLA, CLA or other copyright owner. No further copying (including the printing of digital cuttings), digital reproduction or forwarding is permitted except under license from the NLA, www.nla.co.uk (for newspapers) CLA, www.cla.co.uk (for books and magazines) or other copyright body. Article Page 2 of 8 A23797 - 3 Media: British Journal of Nursing (BJN) {Main} Edition: Date: Thursday 12, November 2020 Page: 58,59,60,6 Table 2. Falls interventions 1. Ensure that the call bell is within reach, is working and that the patient can use it (consider other ways of communication if required) 2. Ensure that footwear is non-slip, low heeled and well-fitting. Check hospital slipper socks daily for grip. Check feet, refer any problems to podiatrist 3. Ensure that spectacles are within reach and are clean 4. Urinalysis to be taken on admission. Respond to requests for toilet facilities with urgency—within no more than 5 minutes (consider communication needs) 5. Ensure the correct use of hearing aids 6. Assess needs/ensure walking aids are appropriate and within reach at all times 7. De-clutter bed space 8. Ensure that personal belongings are within easy reach 9. Discuss normal activities of daily living with patient and carers 10. Nurse in high-visibility bed 11. Nursing staff to accompany/be within arms reach of patient for high-risk activities such as dressing, toileting, etc 12. Assess frequency of 'comfort/toileting rounds'. Implement toileting chart 13. Request medication review is undertaken on ward round 14. Perform bed rails assessment 15. Consider referral to falls clinic 16. Refer to physiotherapist for range of movement, strength, balance and/or gait exercises 17. Identify risk of falls on handover sheet; ensure that all staff are aware of risk status 18. Check lying and standing blood pressure using manual sphygmomanometer at least three times a week at different times of day. Report any postural drop to medical staff for their review 19. Assess the need for increased supervision/observation—consider one-to- one specialing 20. Implement multidisciplinary visual observation chart 21. Consider low-level bed 22. Assess for suitability for bed/chair alarms o u r b i o l o g i c a l m a k e u p . We m a y g o t h r o u g h life t h i n k i n g that w e m a k e u p o w n m i n d but the reality is that w h e n p e o p l e are free t o d o as they please they o f t e n i m i t a t e e a c h other.' Buchanan, 2007 All t o o often it is these acts of imitation that b r i n g o u r public bodies into disrepute; examples include the Panorama expose of practices at the W i n t e r b o u r n e View care h o m e in Bristol and the Mid Staffordshire N H S Foundation Trust public inquiry (Francis, 2013). W h e n the degradation of patient care spreads through an institution, the team speculated that imitation appeared to be the vehicle. Believing imitation to be a powerful force, it was decided that, rather than taking the usual option of prescribing to staff what they should do through policies and guidelines, the team would focus on h o w individuals should act on a busy ward to reduce the likelihood of their patients falling. Methodology: implementing an emergent design approach As an a l t e r n a t i v e t o t h e p l a n n e d a p p r o a c h used in t h e first year, the project team a d o p t e d a m o d e l referred to as ' e m e r g e n t design'. T h e team a b a n d o n e d the action plans and instead e n c o u r a g e d a m i n d - s e t of curiosity, flexibility and e x p e r i m e n t a t i o n with the n o t i o n that ' n o t k n o w i n g ' is a virtue, not a deficiency. R a t h e r than planning many stages in advance, by using an e m e r g e n t approach, the next step was only planned w h e n the results of the previous were seen.This allowed opportunities to identify and utilise emergent patterns and ideas.The c o n c e p t assumes there is n o 'silver bullet'; small changes are continuously introduced h o p i n g that s o m e will have a ripple effect and encourage o t h e r changes (Suchman, 2011). A n o t h e r similar approach o f ' t r y a lot of things, keep w h a t works' has b e e n f o u n d to be the basis for sustainable success in larger corporations (Collins and Porras, 1994). A well-respected senior nurse was recruited to lead the project. Although she did not see herself as an expert in falls, staff recognised her as a leader with high professional standards, unafraid to challenge the status quo, and with an intellectual curiosity that sought to question and account for the rationale behind the professional actions of all involved in the delivery of patient care. Multifactorial complex problems such as falls have a tendency to be a challenge w h e n applying the dominant approaches to quality i m p r o v e m e n t , such as care bundles, guidelines, educational programmes, action plans, etc. Instead, the project team acknowledged that n o o n e individual, the project lead included, held all the answers to r e d u c i n g the Trust's falls rate. Utilising the Trust's online incident reporting system, the senior nurse augmented the incident investigation process to include an after action review (AAR) debrief following each reported fall.The A A R process facilitates reflection and feedback o n individual, team and collective p e r f o r m a n c e , in a n o n - confrontational fashion, with the spirit of an A A R being o n e o f l e a r n i n g not blame. Schindler and Eppler (2003) suggested that creating an atmosphere that is conducive to encouraging openness and trust are crucial goals of the process to enable team learning, team building and team integrity. An A A R is based around four overarching questions: • W h a t was expected to happen? • W h a t actually happened? • W h y was there a difference between what was expected and what actually happened? • W h a t are the lessons that can be learnt? T h e process was designed to establish the extent of the deficit between the expectations of those involved and the facts surrounding each adverse event (Morrison and Meliza, 1999; Schindler and Eppler, 2003). In practice, this m e a n t that the senior nurse was alerted immediately, via email, to every fall as soon as it was reported via the Trust's online clinical incident reporting system, a well- established system that had been in use since 2007.This rapid notification fostered m a x i m u m opportunities for immediate learning via the A A R for everybody involved with the care of the patient w h o had fallen. T h e A A R framework provided COPYRIGHT: This cutting is reproduced by Gorkana under licence from the NLA, CLA or other copyright owner. No further copying (including the printing of digital cuttings), digital reproduction or forwarding is permitted except under license from the NLA, www.nla.co.uk (for newspapers) CLA, www.cla.co.uk (for books and magazines) or other copyright body. Article Page 3 of 8 A23797 - 3 Media: British Journal of Nursing (BJN) {Main} Edition: Date: Thursday 12, November 2020 Page: 58,59,60,6 a structure for talking about what was expected in relation to how to act in a way that minimised the chances of the patient falling again and future patients from falling. At the same time, the debriefs allowed the opportunity to review the difference between what was expected and what actually happened. T h e expectation-reality deficit was identified for each patient fall and the ward staff were given the opportunity to discuss and reflect on how events might have been prevented and h o w local systems or patterns of working could be improved. T h e emphasis was on constantly improving the learning about falls in each specific local context with the staff involved in caring for the patient that had fallen, thus empowering staff to affect change rather than portraying the process as punitive. T h e focus was exclusively on the local context of the ward— its layout, patients, staffing and leadership were critical. T h e senior nurse was also mindful of generalising the results to a wider context. For example, w h e n intentional toileting made a difference on o n e ward, the results were shared but not imposed on other wards. T h u s through continuous AARs, wards learnt h o w to manage the risk of patients falling on their individual ward. T h e A A R debriefs were with the staff directly involved in the care of the patient w h o had fallen (and when possible also the patient) and focused on exploration of their expectations of the o u t c o m e of their actions prior to the patient falling. In some cases, they had not anticipated the risk of a fall; in others, they had been forced to make pragmatic choices between other care priorities for patients, which had been judged to be of greater importance at the time. An important finding was that in these group reflections, the senior nurse moved beyond the more conventional use of the A A R tool as solely a technique for identifying a 'root cause' of the fall, to engaging in reflexive acts of imagination and exploration of future consequences and creative challenge to local constraints and habits such as 'the way it is around here' or'we've always done it like this.' Learning was generated by working with frontline staff in developing safety solutions, which resulted in the continual updating of mutual expectations around nurses' behaviour in relation to falls-aware practice. Instead of being told what to do solely through policies and guidelines, it was contended that it is the staff delivering the care w h o hold the answers to improving the safety of care. At that time, the project team's belief was that behaviours spread through social groups by imitation, thus, if one nurse (particularly, an influential nurse) acts in a certain way, this increased the likelihood that others will follow this individual's behaviour through role modelling (Buchanan, 2007). As this approach has focused on behaviours (and behaviours are habitual), the team contended (as evidenced by the consistent falls rate in recent years) that because habits are hard to break, sustainability is built into the programme for 2 the reasons explained below. Staff were encouraged to develop practical ideas on how they | would prevent the recurrence of falling. All ideas were embraced < and tested through constant measurement of the wards' falls | rates. T h r o u g h countless iterations, the expectations around o falls-aware practice evolved. An example of this would be: • A patient with a high falls risk should have a rapid response when they use the call bell.This becomes a habit • T h e patient still falls as they are left alone in the toilet • T h e patient's call bell is answered rapidly and a member of staff waits for the patient while they are in the toilet in order to accompany the patient back to bed. This challenged the notion of 'best practice' as something fixed, towards it b e i n g u n d e r s t o o d as s o m e t h i n g that is continuously evolving and being re-tested in a local context. It also highlighted the idea that'best practice'is not a simple rule to follow, devoid of the need for re-evaluation and consideration of the application of the rule locally. For example, this might mean weighing up the competing values such as 'patient safety' with others such as 'maintaining dignity' w h e n choosing whether to remain present while the patient uses the toilet. If a familiar practice delivers an outcome that is unexpected or detrimental, it makes sense to re-evaluate whether or not it is still the right thing to do. If not, an alternative emerges through acts of group reflection in the A A R . If that worked, we found the revised practice becomes habituated because it then seemed the obvious thing to do. This is in line with the pragmatic tradition that underpins the team's approach (Stacey and Mowles, 2016). T h e continued focus on expectations, and the evolution of these expectations in conjunction with the immediate discussion and re-examination of the project team's expectations of'falls- aware' behaviour meant that learning from each specific fall was immediately translated into practice rather than through a written action plan to be enacted at a later date. W h e n talking about this work, the team is often asked for practical examples of what they do. T h e policy is that all patients are risk assessed on admission for any k n o w n risk factors that may precipitate a risk of falls. These include age, a history of falls, mobility, level of comprehension, any sensory impairment, medication, sleep patterns, neurological problems and continence. O t h e r factors such as alcohol abuse, anticoagulant therapy, postural hypotension, recent surgery, sepsis or epidurals are also taken into account. T h e assessment chart includes practical suggestions (see Table 2) for how staff can mitigate the risks of their patients falling, ranging from practical issues such as ensuring the bedside area is clutter-free, through to measures such as the need for increased supervision and observation, consideration of one-to- one specialing, (enhanced patient supervision by nursing staff) and the implementation of the Trust's multidisciplinary visual observation chart. Consideration is also given to the use of a low-level bed and the patient is assessed for suitability for a bed or chair alarm. In the event of a patient falling, an algorithm is followed for post-fall m a n a g e m e n t and care. Each fall is reviewed to ensure that lessons are learnt, as outlined above. This may include further revision of both the risk assessment tool and the falls action care plan for that patient. T h e focus was on exploring each fall and finding solutions within the existing resources. As the falls rates fell, the team asked staff what had made the most difference. O n e replied: 'I don't know, but preventing falls just s e e m s m o r e at the front o f m y mind.' COPYRIGHT: This cutting is reproduced by Gorkana under licence from the NLA, CLA or other copyright owner. No further copying (including the printing of digital cuttings), digital reproduction or forwarding is permitted except under license from the NLA, www.nla.co.uk (for newspapers) CLA, www.cla.co.uk (for books and magazines) or other copyright body. Article Page 4 of 8 A23797 - 3 Media: British Journal of Nursing (BJN) {Main} Edition: Date: Thursday 12, November 2020 Page: 58,59,60,6 Table 3 . Falls r a t e for t h e e i g h t w a r d s involved in t h e s e c o n d falls project Ward April 2 0 0 9 — March 2 0 1 0 April 2010—-March 2 0 1 1 % difference in falls rate 2 0 0 9 - 2 0 1 0 to 2 0 1 0 - 2 0 1 1 Reported falls Falls rate per 1 0 0 0 bed days Reported falls Falls rate per 1 0 0 0 bed days % difference in falls rate 2 0 0 9 - 2 0 1 0 to 2 0 1 0 - 2 0 1 1 Ward 1 3 2 5.3 3 2 5.3 -0.3 Ward 2 75 11.5 36 5.0 -56.3 Ward 3 28 3.3 4 2 5.3 60.6 Ward 4 104 16.2 59 8 . 0 -50.5 Ward 5 39 4.3 6 0 7.3 4 7 . 2 Ward 6 5 1 8.2 4 2 5.7 -30.6 Ward 7 5 4 9.57 43 6.9 -28.3 Ward 8 103 16.3 63 9.4 -42.5 Total 4 8 4 9 . 1 3 7 6 6.6 -27.5 It was also noticeable that staff were beginning to behave differently, for example, asking if a patient was high risk before they were transferred into their ward and automatically putting them in a high-visibility area if they were. Or, accompanying the patient to the toilet and waiting for them rather than leaving them alone in the bathroom. These subtle shifts in behaviour and in how staff were talking about falls have, the authors argue, collectively made a big difference to the falls rate, as outlined in the results below. Results In the second year, as well as changing the theoretical approach to the project, the team also curtailed its ambition by focusing on one geographical block that housed eight elderly care and medical wards.This block was chosen as it accounted for almost a third of the falls in the Trust and the wards had some of the highest falls rates in the hospital. D u r i n g the first year, these eight wards reported 484 falls at a rate of 9.1 falls per 1000 bed-stay days. In the second year, the rate of falls was reduced by 29% to 6.6 falls per 1000 bed stay days. As Table 3 illustrates, six of the eight wards reduced their falls rate during 2010-2011. During 2011-2012, the B S U H project expanded to focus on reducing the rate offalls across an additional 31 acute inpatient wards, which between 2011 and 2019 accounted for 84% of the Trusts total number offalls. Areas excluded from the programme included intensive care, maternity and paediatrics as these areas had a relatively low rate of falls. Accident and emergency was also excluded as it does not have beds and therefore it was not possible to calculate a rate. Similarly, outpatient clinics, discharge lounges, dialysis units and day case areas were also excluded. Table 4 and Figure 2 illustrate that the annual falls rate has continued to decline, with a rate of approximately 3.4 falls per 1000 bed days being sustained over the past 5 years (despite the absence of the senior nurse w h o left the Trust in 2016). By the end of the year 2018—2019, the overall reduction in falls was 46%, based on the rate in 2010-2011 o f 6 . 2 3 falls per 1000 bed days. Based on costings from N H S Improvement, the 5108 fewer falls between 2011 and 2019 saved theTrust ^ 1 3 . 3 million. Limitations It should be noted that between 2009 and 2018 the Trust opened new wards and closed some services. For the purpose of consistency, analysis in this article only contains data on those wards that remained unchanged.The project team was conscious that the measurement of the impact of this initiative was reliant on self-reported data (clinical incidents) .Throughout the initiative the team constantly monitored the distribution of reported falls each day using the Poisson distribution. This approach allowed the team to calculate the probability of a certain number of reported falls on any particular day, while comparing this to the actual number of reported falls. Over the duration of this project course compliance or fit against the Poisson distribution curve rose from 86.8% in 2009-2010 to 94.2% for the wards involved in the second falls project. A finding that suggests the focus on falls did encourage a greater level of reporting. Table 4 . Falls r a t e a n d number of falls 2 0 0 9 - 2 0 1 9 Year Reported falls Bed days Falls rate per 1 0 0 0 bed days Cumulative % difference since 2 0 1 0 - 2 0 1 1 Number of falls saved (based on rate of 6 . 2 3 ) Financial saving 2 0 1 1 - 1 2 1 4 0 1 2 4 2 5 5 3 5.78 -7.27 1 1 0 2 8 6 2 7 3 2 0 1 2 - 1 3 1 3 5 6 2 4 9 4 9 7 5.43 -12.75 1 9 8 5 1 5 7 5 2 2 0 1 3 - 1 4 1 0 8 8 2 6 3 2 8 5 4.13 -33.66 5 5 2 1 4 3 5 8 9 0 2 0 1 4 - 1 5 8 9 5 2 5 7 4 5 4 3 . 4 8 -44.19 7 0 9 1 8 4 3 2 4 0 2 0 1 5 - 1 6 8 9 6 2 6 1 8 2 7 3.42 -45.06 7 3 5 1 9 1 1 4 7 4 2 0 1 6 - 1 7 877 2 6 1 6 0 3 3 . 3 5 -46.18 7 5 3 1 9 5 7 2 4 5 2 0 1 7 - 1 8 8 4 5 2 5 2 8 1 9 3 . 3 4 -46.34 7 3 0 1 8 9 8 1 6 2 2 0 1 8 - 1 9 8 6 2 2 5 5 0 9 1 3 . 3 8 -45.75 727 1 8 9 0 7 6 4 Total 8 2 2 0 2 0 4 4 1 2 9 5 1 0 8 1 3 2 8 0 5 3 9 COPYRIGHT: This cutting is reproduced by Gorkana under licence from the NLA, CLA or other copyright owner. No further copying (including the printing of digital cuttings), digital reproduction or forwarding is permitted except under license from the NLA, www.nla.co.uk (for newspapers) CLA, www.cla.co.uk (for books and magazines) or other copyright body. Article Page 5 of 8 A23797 - 3 Media: British Journal of Nursing (BJN) {Main} Edition: Date: Thursday 12, November 2020 Page: 58,59,60,6 Discussion It was evident at the early stages of this initiative that there was a generalised attitude and belief that falls were largely unavoidable and therefore it was accepted that patients may fall while in hospital.The work by the project team and nursing staff over the past 10 years has reinforced the team s belief that understanding how such nursing behaviours and beliefs arise is the key to reducing patient falls. T h e initial hunch about the importance of imitation and positive role modelling in facilitating falls- safe behaviour and habits to spread through the Trust's ward- based teams was a helpful starting point, but a research grant and subsequent study enabled a more detailed understanding about the processes involved in changing professional values and norms regarding falls-safe practice (Norman et al, 2015). T h e project team's findings built on and developed the initial assumption (informed by change and educational theories worked with over the years), that people learn through mimicking others more experienced or powerful than themselves, and that knowledge is something that is transferred from an expert to a novice, led to a different understanding of behaviour change, drawing on the pragmatic tradition (Mead, 1934). It became evident from the team's research that there was much more going on in the complex social interaction between the senior nurse and staff than 'role modelling' or simple mimicry. By adopting the A A R framework in the way described, where reflexive conversation, rather than completion of the clinical incident investigation f o r m (previously normally completed by the most senior staff for the sake of expediency, if at all), was the main focus, the project team recognised that the practising nurses and healthcare assistants often held the solution to many complex safety and quality problems.Thus the specialist expertise of the senior nurse involved the skilful facilitation of the debriefing, providing reflexive space in which she and others were able to have potentially difficult conversations about w h o had done what and why. The project team recognised that such conversations had often been avoided, because it was felt that they may cause conflict and provoke strong emotions, especially if the patient has sustained a serious injury. Staff reported feeling guilty, ashamed, angry and blamed—both with themselves and with each other.They often found it hard to talk about such feelings, both because of fear of being blamed themselves, or of upsetting others by calling their professional judgement and practice into question.This is consistent with pragmatist John Dewey's definition of reflection as 'the painful effort of disturbed habits to readjust themselves' (Dewey, 1922; Brinkmann, 2013).Thus as our habits become problematic in each local context, we formulate conscious ideas that are tested through active action, ie in this way of thinking, thought is action, rather than something that precedes it, as assumed in the action-planning activities described earlier.The A A R debriefing provided the opportunity for the ward team to 2 develop a shared understanding about possible causes for each 3 fall, and renegotiate together whether their habitual practices 1 (norms) for falls prevention were fit for purpose in their specific < local environment. In this way of understanding, learning arises § from their shared reflection about a specific incident, after which © it no longer makes sense to follow previous habits of practice. Figure 2. Rate of inpatient falls per 1 0 0 0 bed days 2 0 0 9 to 2 0 1 8 This is because staff expectations about the possibilities of the outcomes of their actions/inaction are no longer the same, due to the insights gained from hearing other staff's perspectives about what happened w h e n a patient fell, why they all did what they did, and the unforeseen consequences of their own actions and the actions of others. Although initially anxiety- provoking for all concerned, as the debriefings became the norm and valued by those participating, something seemed to shift, with staff coming to see themselves as having the potential to influence the emergent patterns of practice in their ward, both by their actions and omissions. Subsequent research (Norman et al,2015), helped the project team to recognise that what also shifted fundamentally was how staff were talking more frequently about preventing falls, and also how they were talking about this topic. As Shaw noted: 'Conversation itself is the key process through w h i c h f o r m s o f organising are dynamically sustained and changed.' Shaw, 2002 T h e project team concluded that the capacity of the ward teams to keep alive this narrative t h e m e and its o n g o i n g evolution may be one of the main reasons for sustaining the lower falls rate after the senior nurse leader left the project. T h e subsequent wider conversations between wards in this growing community of practice identified generalised themes about what constituted falls-safe practice and these were then re-tested and adjusted once again in each specific ward situation. As one clinical manager noted in one of our focus groups when asked what had changed their mind: 'So what's c h a n g e d is that, rather than saying, "Okay we've had a fall last night, okay, is she alright? Yes, she is fine." [Instead] it's, "You've had a fall; have y o u d o n e the after action review? Yes? What was the learning? What have y o u d o n e now? What actions have y o u put in COPYRIGHT: This cutting is reproduced by Gorkana under licence from the NLA, CLA or other copyright owner. No further copying (including the printing of digital cuttings), digital reproduction or forwarding is permitted except under license from the NLA, www.nla.co.uk (for newspapers) CLA, www.cla.co.uk (for books and magazines) or other copyright body. Article Page 6 of 8 A23797 - 3 Media: British Journal of Nursing (BJN) {Main} Edition: Date: Thursday 12, November 2020 Page: 58,59,60,6 place to r e m o v e that?" "Well, I've m o v e d that patient into a high visibility bed." Or: "Have y o u put the falls alarm on?" It's just linking it all, in a m o r e intelligent way, rather than, "What a s h a m e , is she okay? Is she fine? Okay, well then, fine," and just m o v i n g o n , rather than learning f r o m that.' T h e manager also said: 'And I think that learning c o m e s over time, y o u don't instantly learn ... after o n e after action review, it's that k n o w l e d g e we're building o n a ward, within that c o m m u n i t y , to allow t h e m to see things and prevent t h e m before they happen ...You know, what's been done, it's like, instinct, it's a n o r m a l reaction for this to the i n f o r m a t i o n , this is what we've d o n e , and this is what we're g o i n g to d o to prevent it. I just think w e are m o r e intelligent.' So rather than seeing 'best practice' as something located 'in' an expert w h o empowers others, or a policy or guideline, staff now see this as an ideal, a value, which is constantly tested in practice, which may require further investigation and need revising when this fails to deliver what was expected or they are surprised in some way. T h e project team also learnt that nurses' knowledge of falls prevention was seldom the problem in terms of falls awareness w h e n assessing and documenting the risk of patients falling.This helped explain why approaches such as ensuring staff are aware of policies and guidelines while having some impact, on their own failed to yield the more radical improvements shown from year 2 onwards. T h e real challenge arose when staff had to make contextual judgements about falls-safe practice in complex situations, with competing priorities, in which every patient and each ward setting is slightly different and best practice is more nuanced than simply following the protocol, as in the above example, where beliefs about patients' dignity and privacy conflicted with those of safety when deciding whether or not to stay with a frail patient w h o was using the toilet. Conclusion Rather than now seeing falls solely as either a 'systems' problem (system failure), or one that locates the cause of the fall with individual staff (poor performance or lack of competency), as informed by the traditional improvement methods taught at the start of the project in 2009, the Trust's falls reduction programme drew on insights from complexity science and its helpful concept of emergence. In this way of thinking, a fall is understood as an event that emerges out of the micro- interactions between nurses and their patients, that is, because of what everyone is d o i n g / n o t doing. From this perspective, relationships are fundamental and falls emerge from complex interactions, and not in isolation.This understandi