Introduction Hip fractures are associated with high rates of morbidity and mortality and their incidence is set to increase. The National Hip Fracture Database and the Best Practice Tariff were introduced with the aim of improving patient care. This complete audit cycle charts the substantial clinical improvements that were achieved in a busy district general hospital.
Methods The first audit cycle comprised 379 patients who were admitted between May 2012 and April 2013. The primary audit criterion was operative intervention within 36 h of admission. Variation according to the day of the week of admission was assessed to evaluate specific deficiencies in local service provision. The principle audit intervention was the introduction of two additional morning trauma lists. A re-audit of 162 patients was conducted prospectively between January 2014 and June 2014.
Results Mean time to theatre was 49±39 h during the first audit cycle compared with 27±19 h (p<0.0001) during the second. Consequently, the proportion of patients undergoing operative intervention within 36 h of admission improved from 41% to 78% (p<0.0001). Overall achievement of Best Practice Tariff was significantly higher during the second cycle: 28% vs 73% (p<0.0001).
Conclusions Significant improvements in the quality of hip fracture care were achieved following this audit. These were accomplished by rigorously analysing the variation in Best Practice Tariff achievement according to the day of the week on which patients were admitted. Targeted interventions could therefore be introduced that addressed specific problems in local service provision.
- Healthcare quality improvement
- Quality improvement
Statistics from Altmetric.com
Hip fractures present a major challenge to orthopaedic trauma and rehabilitation services. The incidence in the UK is rising annually and is currently estimated at 10.2 per 10 000 per year. There is an exponential increase with age, which is 2–3 times greater in women.1 The number of patients with a hip fracture in the UK is therefore estimated to rise, given the aging population, from ∼75 000 patients per year currently to 91 500 in 2015 and 10 100 in 2020.2
Patients with a fractured neck of femur often have a plethora of comorbid conditions and are at high risk of developing complications.3 Overall, 1-year mortality rates are high and vary between 10% and 37%.4–6 Mortality is even higher in those patients residing in a nursing home at the time of injury.7 Approximately 25% of patients are purported to require a higher level of long-term care, and those who are discharged to the community often have a decline in mobility level and require assistance with their activities of daily living.8 ,9
There is established evidence that organised and coordinated multidisciplinary team (MDT) management improves outcomes in patients with a hip fracture.3 Particularly important is early operative intervention, with improved patient survival resulting from expeditious surgery.5 ,10–13 In addition to a reduction in morbidity and mortality, prompt surgical intervention significantly decreases postoperative length of hospital stay.14
The Best Practice Tariff (BPT) was introduced by the Department of Health in April 2010. It formalises a care pathway reflecting the factors known to influence outcome in patients with a hip fracture. It financially rewards hospitals when optimal care is delivered and is determined by the achievement of certain targets (box 1). Compliance is monitored through the National Hip Fracture Database (NHFD).15
Best Practice Tariff criteria for patients with a hip fracture
(a) Time to surgery within 36 h from arrival in an emergency department, or time of diagnosis if an admitted patient, to the start of anaesthesia
(b) Admitted using an assessment protocol agreed by geriatric medicine, orthopaedic surgery and anaesthesia
(c) Assessed by a geriatrician in the perioperative period (within 72 h of admission)
(d) Postoperative geriatrician-directed multiprofessional rehabilitation team
(e) Fracture prevention assessments (falls and bone health)
(f) Two Abbreviated Mental Tests performed, with the first test carried out before surgery and the second after surgery but within the same spell
The Royal Bolton Hospital is a busy district general hospital located in the North West of England which serves a population of approximately 300 000 people. The mean (±SD) annual number of patients admitted with a hip fracture is 353±21, compared with the regional mean of 331. However, performance with regard to the BPT was poor. Operative intervention within 36 h of admission was the main limiting factor, with Bolton one of the poorest performing hospitals in the country. Only 41.4%, compared with the national average of 71.4%, underwent operative intervention within 36 h of admission,15 and, as a result, only 28% of patients qualified for BPT. An improvement in the hip fracture service was required.
Previous studies have used the BPT criteria as an audit tool to examine hip fracture care. Patel et al16 described an improvement of 12% in patients undergoing operative intervention within 36 h following a change in priority of patients on their trauma lists. Further, Khan et al17 and Britton and Nash18 demonstrated a 25% and 29% improvement, respectively, following the addition of an extra trauma operating session.
The aim of this audit was to improve the care of patients with a fractured neck of femur. The primary outcome measure was the achievement of operative intervention within 36 h of admission. Secondary outcomes were the attainment of the remaining BPT criteria. To maximise effectiveness, intervention ought to be targeted and should specifically address local deficiencies in service provision. This requires interrogation of the admission profile of patients with a hip fracture against trauma list availability. While previous studies have analysed the impact of additional trauma sessions, we are not aware that a rigorous examination of local deficiencies has been conducted to enable targeted intervention.
Data were exported from the NHFD after approval to access and export the results was granted by the host institution. The database is prospectively populated by a team of trauma coordinators. Exclusion criteria were: (1) patients younger than 65 years; (2) patients with pathological or non-fragility fractures; (3) patients treated operatively elsewhere but transferred for rehabilitation. These criteria, extrapolated from the NHFD, were principally designed to exclude younger patients who sustained a hip fracture following a high-energy injury. The criteria were only met in a very small number of cases and in each case it was clear from the initial assessment. The data collected included demographic characteristics, time and day of admission, time from admission to theatre and orthogeriatric assessment, and achievement of the BPT criteria. This work met the criteria for operational improvement activities and is therefore considered exempt from ethics review.
The first audit cycle was conducted retrospectively in July 2013 and comprised 379 patients who were admitted during a 1-year period between May 2012 and April 2013. The time from admission to theatre was analysed according to the day of the week that a patient was admitted. Problem days for achieving the 36 h time to theatre target were identified and compared with trauma list availability. Patients admitted on a Tuesday, Wednesday or Thursday underwent surgery within 36 h of admission in only 23%, 28% and 27% of cases, respectively (table 1). (The baseline results are reported fully below.) The audit interventions were formulated after involvement of the orthopaedic clinical lead, divisional head and the orthogeriatric service. The targeted intervention was the introduction of two additional trauma sessions on a Monday and Wednesday morning. A hip fracture escalation plan, which established a precedent for the cancellation of elective procedures, was also introduced. In addition, the orthogeriatrician attended the morning trauma meeting. This was designed to improve team working and streamline both orthogeriatric and anaesthetic assessment, allowing improved utilisation of the trauma lists. A 6-month period was afforded to allow the changes to become established into departmental working practice. A re-audit completed in July 2014 was conducted prospectively during a 6-month period between January 2014 and June 2014 and included 162 patients.
Patients presenting with a neck of femur fracture were admitted under the care of trauma and orthopaedics. Patients who sustained a fracture following a fall while already an inpatient under another speciality were either transferred to an orthopaedic ward before surgery or had their care taken over after surgery. A total of 21 patients (4%) sustained a fracture while already inpatients. All patients were reviewed by a full-time orthogeriatrician and received ongoing regular input as necessary.
The appropriate descriptive statistics are given, which include mean, SD and percentage. Achievement of BPT criteria was calculated by dividing the number of satisfactory episodes by the number of eligible episodes within the given time frames. The BPT criteria were treated as categorical variables and compared between audit cycles using a two-sample t test between percentages. Length of stay and time for admission to theatre and orthogeriatric assessment were compared between audit cycles using a Student's t test. Analysis of variance was used to compare data between days of admission. p values <0.05 were accepted as significant.
Figure 1 outlines the time at which patients with a hip fracture were admitted. The number of patients admitted rose steadily during the morning until lunchtime. Subsequently, this plateaued during the afternoon, before declining into the evening and overnight. All patients having surgery on the day of, or the day after, admission fall within the 36 h BPT criterion. However, there is an opportunity for patients admitted towards the end of a day to undergo surgery on their second full day of admission and still fall within this 36 h window. This is illustrated in figure 2, which shifts the admission profile forwards to reflect the expiry of the window. The increased value of a morning trauma list over an afternoon or evening trauma list is demonstrated. In Bolton, a morning trauma list would have enabled an additional 90 patients to undergo surgery while on their second full day of admission and still fall within the 36 h window. This compares with an additional 38 patients for an afternoon trauma list or 24 patients for an evening trauma list.
Trauma list provision
The trauma list availability at the Royal Bolton Hospital is shown in figure 3. Hip fracture operations are performed on all trauma lists, but specialist upper limb trauma is incorporated on the days indicated. The two additional trauma lists were introduced on a Monday and Wednesday morning, reflecting the additional benefit of a morning list (see ‘Admission profile’ section). This was to compensate for the upper limb trauma sessions, while falling within what was possible logistically within the general operation of the department.
Time to theatre
The percentage of patients undergoing operative intervention within 36 h of admission significantly improved following the audit intervention: 41% vs 78% (p<0.0001). Table 1 and figure 4 illustrate the improvement according to the day of the week that patients were admitted. Mean (±SD) time to theatre was significantly shorter in the second audit period: 49±39 h vs 27±19 h (p<0.0001). Considering the first audit cycle, there were no significant differences in time from admission to theatre according to day of admission (p=0.0689–0.8746). With regard to the second cycle, time to theatre was significantly longer for those patients admitted on a Tuesday compared with a Wednesday (p=0.0340), Thursday (p=0.0246) and Friday (p=0.0101).
Time to orthogeriatric assessment and remaining BPT criteria
The proportion of patients undergoing orthogeriatric assessment within 72 h of admission significantly increased in the second audit period: 77% vs 95% (p<0.0001). During the first audit cycle, time until orthogeriatric assessment was significantly longer for those patients admitted on a Friday (74±114 h) compared with a Monday (33±23 h; p=0.001), Tuesday (25±23 h; p<0.0001), Wednesday (26±19 h; p<0.0001), Thursday (34±32 h p=0.0002), Saturday (46±13 h; p=0.0051) or Sunday (29±17 h; p<0.0001). A similar pattern was seen during the second audit cycle.
Significant improvements with regard to the use of an assessment proforma, falls assessment and pre- and post-operative Abbreviated Mini-Mental Test were made during the second audit period. The rates of review of bone protection medication and MDT assessment were very high during both audit cycles (table 2).
Length of stay
Total length of stay was significantly reduced in the second audit period: 18±12 days vs 15±10 days (p=0.0027). In addition, significant reductions in postoperative length of stay were seen in the second cycle: 16±12 days vs 14±10 days (p=0.0272).
Hip fractures present a substantial challenge to orthopaedic trauma services, and the incidence is set to rise with the changing population demographic.1 ,19 Patients often have multiple comorbidities, and mortality rates are high.4 Optimal management requires a coordinated approach from multiple health professionals, including orthopaedic surgeons, orthogeriatricians, anaesthetists, and nursing and rehabilitation staff. Indeed, it is clearly established within the literature that timely surgical intervention and MDT assessment can improve patient outcomes.3 ,11 ,12 ,18
The BPT formalises a number of targets that reflects optimal care, and a financial reward is provided for hospitals that satisfy all criteria. The care of patients has been shown to improve following the introduction of the BPT.20 The NHFD is a national audit project that was introduced in 2007. The aim of this was to drive sustained management improvement by generating continuous and comparative data. A hospital's BPT achievement is based on data extracted from the NHFD. It was clear, after a local audit of BPT compliance in July 2013, that the hip fracture service in Bolton needed improvement: only 41% of patients underwent operative intervention within the 36 h target and only 28% met all BPT criteria.
A delay in time to theatre was the biggest limitation in achieving the BPT, as illustrated by the close correlation between the variables in figure 5. Effective targeted interventions were needed to ensure that substantial improvements were achieved. A rigorous analysis of the hip fracture admission profile was undertaken after the first audit cycle and was compared with the daily trauma list provision. This enabled the local deficiencies in service provision to be identified. Achievement of operative intervention within 36 h of admission was lowest for those patients admitted on a Tuesday (23%). This occurred because specialist upper limb trauma was undertaken on a Wednesday and there was no Thursday morning trauma list available. Consequently, the intervention was the addition of two morning trauma sessions (on a Wednesday and Monday mornings) (figure 3). While these were introduced to offset upper limb trauma, they were required to fit around other service provisions. Morning trauma lists are more effective than afternoon and evening sessions, as a higher number of patients can undergo operative intervention on the second full day of admission and still fall within the 36 h target. In addition, a hip fracture escalation plan (figure 6) was developed which provided a mechanism for cancelling elective surgery when no trauma operative space was available.
A re-audit was undertaken in July 2014. Time to theatre had dramatically improved from 49 to 27 h (p<0.0001) and the proportion of patients satisfying the 36 h BPT criterion had significantly improved by 37% from 41% to 78% (p<0.0001). Previous audits of hip fracture management against BPT criteria have shown improvements ranging from 12% to 29%, but we are not aware of a bigger progression within such a short time frame.16–18 This illustrates the effectiveness of targeted interventions that can be introduced when both the local admission profile and any problems with service provision are fully understood. A full analysis of the impact of increasing trauma service provision would require a concurrent assessment of compliance with the 18-week elective target. This formal analysis was not undertaken as part of the work, but anecdotally there has not been a negative impact.
Patients admitted on a Friday waited significantly longer for an orthogeriatric assessment, which was to be expected given that there is no weekend service. Overall, there was a significant improvement of 18% (to 95%), in patients undergoing orthogeriatric review within 72 h of admission. Significant improvements in the use of an admission proforma, falls assessment and pre-and post-operative Abbreviated Mini-Mental Test reflects this. Interestingly, significant improvements in the number of patients undergoing operative intervention within 36 h of admission were seen on all days, except Sunday, and not just those related to the introduction of additional trauma lists. It is likely that this is due to the integration of the orthopaedic trauma and orthogeriatric service in the morning trauma meeting and a change in departmental philosophy, with increasing prioritisation being given to patients with a hip fracture. As a result, Bolton, previously one of the lowest performing centres, is now comfortably above the national average of 71%.21
Considering all BPT criteria, there was an improvement of 45%, from 28% to 73% (p<0.0001). This would represent an additional 165 patients eligible for the BPT financial uplift, based on the number of admissions in 2013. This near threefold increase signifies a substantial improvement in patient care. In addition, there is a differential of £1335 paid for achievement of BPT.22 This would translate to a supplementary £220 275 of income to the Trust based on the number of admissions during 2013.
A reduction in both total length and postoperative length of stay was seen during the re-audit period. The total length of stay was reduced by 3 days from 18 to 15 days (p=0.0010). This is likely to reflect both a shorter time to theatre and improved postoperative medical care, with more patients undergoing orthogeriatric review. Thomas et al14 found that an increased preoperative time significantly lengthened postoperative stay, and this work is consistent with his findings.
The reduced length of stay corresponds to an annual reduction of 1214 bed days. Although the cost of a hip fracture bed day is unknown, it has been estimated to be £328.73 for elective knee surgery.23 Translating this figure to patients with hip fractures would correspond to a saving of approximately £400 000. In addition to the financial reward for achievement of BPT, this would constitute a £620 275 financial gain. While a formal cost–benefit analysis is beyond the scope of this work, it illustrates an added benefit to the Trust in addition to improvements in patient care.
The limitations of the work are acknowledged. While the results are limited by the quality of the input into the NHFD, Bolton ranks highly in terms of data completeness according to the 2013 NHFD report. Unfortunately, reliable readmission data cannot be provided to investigate whether readmission rates are affected by a reduction in length of stay. Finally, the outcome measures used, the BPT criteria, were chosen because data generated from the NHFD is easily accessible, comparable and readily available to all Trusts. However, using mortality rates and health-related quality of life outcomes would provide an alternative perspective on the impact of the audit intervention.
A substantial improvement in the care of patients with a hip fracture has been achieved. This was accomplished by rigorous analysis of variation in BPT achievement according to the day of the week on which patients were admitted. Targeted intervention could therefore be introduced that addressed specific problems in local service provision. It is a simple concept that is translatable to all orthopaedic departments and it could have a wide impact if similar improvements are achieved elsewhere.
Contributors DH: responsible for data analysis and interpretation and undertook drafting and final approval of the manuscript; accountable for the work. JB: responsible for data analysis and interpretation and undertook manuscript revision and then final approval for publication; accountable for the work. CB and GH: responsible for data acquisition and undertook manuscript revision and then final approval for publication; accountable for the work. JR, AW and PW: responsible for the concept and design of the work and undertook manuscript revision and then final approval for publication; accountable for the work.
Competing interests None declared.
Provenance and peer review Not commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.