Inequalities in Access to Total Hip Arthroplasty for Hip Fracture: … · 2016-04-29 · Confidential: For Review Only Inequalities in Access to Total Hip Arthroplasty for Hip Fracture:

Confidential: For Review O

nly

Inequalities in Access to Total Hip Arthroplasty for Hip

Fracture: Population Based Study

Journal: BMJ

Manuscript ID: BMJ.2015.028371

Article Type: Research

BMJ Journal: BMJ

Date Submitted by the Author: 28-Jul-2015

Complete List of Authors: Perry, Daniel; University of Liverpool, Institute of Translational Medicine Metcalfe, David; Harvard Medical School, Center for Surgery and Public Health Griffin, Xavier; Warwick Medical School, Warwick Orthopaedics Costa, Matthew; University of Oxford, Nuffield Department of

Orthopaedics, Rheumatology and Musculoskeletal Sciences (NDORMS)

Keywords: Hip Fracture, Inequalities, Total Hip Replacement, Arthroplasty, Fracture, Osteoporosis, Surgery, Orthopaedics

https://mc.manuscriptcentral.com/bmj

BMJ


nlyInequalities in Access to Total Hip Arthroplasty for Hip Fracture:

Population Based Study

Daniel C Perry, David Metcalfe, Matthew L Costa.

Warwick Clinical Trials Unit, University of Warwick, Coventry, CV4 7AL, UK.

Daniel C Perry NIHR clinician scientist in orthopaedic surgery Institute of

Translational Medicine, University of Liverpool, Liverpool, L12 2AP, UK.

David Metcalfe research fellow Center for Surgery and Public Health, Harvard

Medical School, Boston, MA 02115, USA.

Xavier Griffin NIHR clinical lecturer Warwick Clinical Trials Unit, University of

Warwick, CV4 7AL

Matthew Costa professor of orthopaedic trauma surgery Nuffield Department of

Orthopaedics, Rheumatology, and Musculoskeletal Sciences, University of Oxford,

Oxford, OX3 7HE, UK.

Correspondence to: DC Perry

[email protected]

Daniel Perry designed the study, performed the analysis and drafted the paper.

David Metcalfe contributed to the data analysis, interpretation of results, and

draft manuscript. Matthew Costa and Xavier Griffin contributed to the design of

the study, interpretation of results and critically appraised the paper. All authors

have approved the final manuscript. Daniel Perry is guarantor.

No external funding was received for this work.

Data sharing: Pursuant to the terms of our data sharing agreement with the

National Hip Fracture Database we regret that no additional data can be made

available by the authors.

Page 1 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyAbstract

Objectives To determine whether the decision to offer total hip arthroplasty

(THA) to individuals with a displaced intracapsular fracture of the femoral neck

is based on national guidelines, or if there are systematic inequalities in access to

THA.

Design Observational cohort study using the UK National Hip Fracture Database

(NHFD).

Setting All hospitals that treat adults with hip fractures in England, Wales, and

Northern Ireland.

Main outcome measures Provision of THA to patients considered eligible under

criteria published by the National Institute for Health and Care Excellence (NICE).

Participants All patients within the NHFD that were aged >60 and received

operative treatment for a non-pathological displaced intracapsular hip fracture

between 1st July 2011 and 31st April 2015.

Results 114,119 hip fracture patients were included, 11,683 (10.2%) of which

underwent THA. 32.0% that appeared to satisfy the NICE criteria received a THA,

and only 42.0% of patients that underwent THA actually satisfied the criteria. A

recursive-partitioning algorithm found that the NICE eligibility criteria did not

optimally explain which patients underwent THA. A model with superior

explanatory power drew distinctions that are not supported by NICE: age >77

years and use of a stick for ambulation. Amongst patients satisfying the NICE

eligibility criteria for THA, logistic regression demonstrated that access to THA

was limited based on higher age (OR 0.88, 95% CI 0.87 to 0.88), lower

Abbreviated Mental Test Score (OR 1.44, 95% CI 1.34 to 1.54), higher American

Society of Anesthesiologists (ASA) score (OR 0.74, 95% CI 0.66 to 0.84), male sex

(OR 0.85, 95% CI 0.77 to 0.93), walking with a stick (0.32, 95% CI 0.28 to 0.35),

and quintiles of increasing socioeconomic area deprivation (1.0 (ref) lowest

quintile vs. highest quintile 1.30, 95% CI 1.13 to 1.51). Patients receiving

treatment during the working week were more likely to receive THA than those

at the weekend (OR 1.12, 95% CI 1.04 to 1.21).

Conclusions There are wide disparities in access to THA amongst individuals

with hip fractures and compliance with NICE guidance is poor. Patients with

higher levels of socioeconomic deprivation and requiring surgery at the weekend

Page 2 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyare less likely to receive THA. Inconsistent compliance with NICE

recommendations means that the optimal treatment for older adults with hip

fractures can depend on where and when they present to hospital.

Page 3 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyIntroduction

There are over 70,000 hip fractures in the United Kingdom every year, with a

combined health and social cost of £2 billion.1 Demographic projections estimate

that the annual incidence will increase to over 100,000 by 2020.2 Mortality is

high, with 8.5% of patients deceased within 30 days of hip fracture.3

A number of initiatives have been credited with improving hip fracture outcomes

in the UK.3 The National Hip Fracture Database (NHFD) was established in 2004,

by the British Orthopaedic Association (BOA) and the British Geriatrics Society

(BGS), with the aim of improving hip fracture outcomes through continuous

national clinical audit.4 The NHFD was supported by combined BOA/BGS clinical

guidance5 and later by the Best Practice Tariff for Hip Fracture which rewards

NHS organisations for meeting defined quality standards, including surgery

within 36 hours of arrival at hospital.6 These initiatives have been associated

with improved hip fracture outcomes, including a fall in 30-day mortality from

10.9% in 2007 to 8.5% in 2011.3

Displaced intracapsular hip fractures are at high risk of painful non-union and so

the recommended treatment is either hemiarthroplasty (HA) or total hip

arthroplasty (THA).7-9 In HA, the femoral head is replaced but, in THA, both the

femoral head and acetabulum are replaced. Although the risk-benefit profiles

vary between these two operations, it has been shown that hip fracture patients

undergoing THA have better function and less need for revision surgery than

HA.7 9-11 In June 2011, the National Institute for Health and Care Excellence

(NICE) recommended that THA should be offered to patients with a displaced

intracapsular hip fracture who are “(a) able to walk independently out of doors

with no more than the use of a stick (b) not cognitively impaired and (c)

medically fit for anaesthesia and the procedure”.8 The provision of THA to hip

fracture patients is not explicitly included as a quality indicator within the NHFD

and so the extent to which surgeons comply with this guideline is unknown.

Page 4 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyThis study sought to identify whether the decision to offer THA is based on

factors that are consistent with national recommendations, or if systematic

inequalities exist with regards access to THA for hip fracture.

Methods

An observational cohort study was performed using the NHFD. The study

protocol was approved by the Healthcare Quality Improvement Partnership

(HQIP) but research ethics committee approval was not sought for secondary

analysis of administrative data in line with Governance Arrangements for

Research Ethics Committee (GAfREC) guidelines.12

Data source

The NHFD is commissioned by the HQIP and captures over 95% of hip fractures

treated in England, Wales, Northern Ireland and the Channel Islands. Data

include patient characteristics, fracture pattern, surgical interventions, and

outcomes. These details are typically collected by specialist nurses within each

hospital that provide continuity of care to patients with hip fractures and

manage submissions to the NHFD.

Inclusion criteria

This study included all patients aged >60 years that presented to hospital

between 1st July 2011 and 31st April 2015 with a displaced intracapsular hip

fracture. 1st July 2011 was selected as one month following publication of NICE

Clinical Guideline 124.8 Patients were excluded if their fracture was coded as

“pathological”.

Variables and outcomes

Data cleaning involved several steps. Two patients had ages recorded as >115

years (both >1000 years) which were recoded to exclude this variable. In 27

(0.01%) cases, the Abbreviated Mental Test Score (AMTS) was not recorded as

an integer and so scores were rounded to the nearest integer. On 1st April 2014

the NHFD data collection tool was updated to record mobility differently within

Page 5 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlythe revised database. Earlier data were therefore mapped onto the new version

using an algorithm presented as Appendix A. In the event of hospital trust

reconfiguration (closure/merger), the hospital code at the time of data entry was

used. As a consequence, some hospitals only contributed data for a few months

prior to reconfiguration.

Variables extracted from the NHFD were age (whole years), sex, lower level

super output area (LSOA), date of admission, treating hospital, pre-morbid

mobility, American Society of Anesthesiologists (ASA) physical status

classification score, and AMTS. The ASA score ranges between 1 (healthy

patient) and 5 (moribund patient not expected to survive for 24 hours with or

without surgery). The AMTS is a test of ten questions (e.g. “what is your age?”),

which gives a score from 0 (zero answers correct) to 10 (all correct).

Deprivation scores for patients living in England were determined using the

Index of Multiple Deprivation (IMD) 2007. These scores reflect deprivation

related to income, health and disability, employment, barriers to housing and

services, living environment, education, and crime.13 IMD scores were generated

from LSOAs, which were then categorised into quintiles of deprivation based on

the population of the UK.

Day of the week was determined from the date of admission. In the UK, hip

fracture surgery usually takes place on the next available trauma operating list

which is the day following admission for most patients in the NHFD (>65%).

“Weekend” surgery was therefore identified by admission on a Friday or

Saturday.

Hospital case volume was analysed by decile and defined by the number of

displaced intracapsular fractures admitted to each centre over the study period.

Date of surgery was analysed as seven 6-month periods (1st July 2011 – 31st

December 2015) and one 4-month period (1st Jan 2015 – 31st April 2015).

Page 6 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyStatistical analysis

Guideline compliance was determined using a decision tree ordered to mirror

the NICE recommendations, i.e. based on mobility (mobile outdoors with or

without the use of a stick), cognition (defined as AMTS >8), and fitness for

anaesthesia (defined as ASA 1 or 2).

Recursive partitioning (RP) was used to determine the optimal decision tree that

explains current practice, i.e. to illustrate how the guidelines are being

interpreted. RP is a statistical technique for multivariable analysis that models

how variables are best organised to predict a given outcome (e.g. THA). In RP,

decision trees are built by identifying a variable that best splits the data into two

groups. RP defines a cut-off (split) for continuous or ordinal variables, to enable

the decision tree to correctly classify the maximum members of the population.

Categorical variables are similarly grouped in RP, to build a tree with the least

error. This process is then applied separately to each sub-group and continues

recursively until either a maximum number of steps are reached or no further

improvement is possible.14

RP was undertaken using the “rpart” function in R. The tree was built using 10-

fold cross validation and a negative complexity parameter to ensure that the

maximum tree was built. Predictors included in the model were age, sex,

mobility, AMTS, ASA, IMD quintile, and day-of-the-week of admission. The tree

was pruned using the complexity (“cp”) function of the smallest tree within one

standard error of the best functioning tree, i.e. the tree with the smallest xerror,

which was confirmed graphically. A pragmatic approach was also used to

consider the tree complexity and efficiency related to clinical practice.

Individuals that fulfilled the NICE criteria were further analysed to explore

factors associated with undergoing THA. An RP decision tree was constructed to

model differentiating between THA and no-THA in this subgroup. The treating

hospital was included as a factor variable, which allowed the partitioning

algorithm to select optimal cut-off points for best fit within the model.

Page 7 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyA mixed effects logistic regression model was constructed to explore factors

associated with access to THA amongst patients that fulfilled the NICE criteria.

Age, sex, date of surgery, AMTS, and ASA were included as continuous

predictors; and IMD quintile and weekend surgery as categorical predictors.

Weekend admission was then substituted for day of the week to explore this

predictor further in a second analysis. Hospital case volume was included as a

centre level fixed effect and the unique hospital identifier as a centre level

random effect.

Statistical analyses were performed using R and Stata v.10.0. p<0.05 was

adopted as the threshold for statistical significance.

Results

In the 46-month period between 1st July 2011 and 31st April 2015, 248,013 hip

fracture patients were recorded within the NHFD. 114,119 (46.0%) satisfied the

study criteria as they were aged >60 years and sustained a non-pathological

displaced intracapsular hip fracture. Within this group, 11,683 (10.2%)

underwent THA. However, only 32.0% of the 21,193 that satisfied the NICE

criteria received a THA (Figure 1) and, of the 11,683 patients that did undergo

THA, only 4,903 (42.0%) satisfied the criteria.

The RP algorithm identified ten terminal nodes (nine splits) as the most

predictive model, although this offered little improvement over five terminal

nodes (four splits) (Figure 2). The variable with the greatest importance was

patient age, with a cut-off of at 77 years defining the initial split (Figure 3). The

mobility split occurred between patients that ambulate independently and those

that required the use of a stick. The other important predictive variables were as

recommended by NICE, with splits occurring as predicted at ASA>3 and AMTS>8.

Using the decision tree, the unexplained variation in practice across the whole

dataset reduced from 12.9% (NICE guidelines) to 9.4% (recursive model).

Page 8 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyAmongst the 21,193 patients fulfilling the NICE eligibility criteria, the RP

algorithm identified 20 terminal nodes (19 splits) to be the most efficient, though

after three splits (four terminal nodes) the complexity of the tree increased

markedly with little associated gain in efficiency (Figure 4). Age was again the

most significant predictor, with 79 years identifying the splitting point (Figure 5).

For patients aged 79 years and above, the treating hospital was the next most

important predictor (further details in Appendix B), followed by mobility (with

or without the use of a stick). Hospital variation amongst individuals fulfilling the

NICE guidelines was considerable (Figure 6). 77% of the variation in practice

could be explained using this RP algorithm.

Date of surgery demonstrated that there was a progressive increase in the

provision of THA for eligible individuals over the study period (Table 1).

Table 1: The proportion of eligible patients undergoing THA by time period.

Period Total number of individuals

undergoing THA / Total number of

individuals fulfilling NICE criteria.

Percentage

undergoing

THA

1st July 2011 – 31st Dec 2011 453/ 2,020 22%

1st Jan 2012 – 30th June 2012 649/ 2,409 27%

1st July 2012 – 31st Dec 2012 804/ 2,703 30%

1st Jan 2013 – 30th June 2013 942/ 3,041 31%

1st July 2013 – 31st Dec 2013 1,007/ 3,099 32%

1st Jan 2014 – 30th June 2014 1,104/ 3,077 36%

1st July 2014 – 31st Dec 2014 1,160/ 3,094 37%

1st Jan 2015 – 30th April 2015 661/ 1,089 38%

The logistic model (Table 2) showed that hospital volume declines did not effect

THA provision (OR 1.02, 95% CI 0.97 to 1.08). However, increasing age (0.88,

0.87 to 0.88), AMTS (1.41, 1.31 to 1.54), and ASA (0.74, 0.66 to 0.84) were

associated with fewer THAs, as was male sex (0.85, 0.77 to 0.93). Admissions

during the working week (Sunday-Thursday) had the highest rates of THA

provision (1.12, 1.04 to 1.21). There was a stepwise decrease in THA surgery

Page 9 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlywith increasing area level deprivation so that the most deprived quintile had

least access to THA (1.30, 1.13 to 1.51).

Table 2: Mixed effects logistic model to demonstrate those receiving a THA

amongst those deemed eligible using the NICE guidelines.

Variable OR (95% CI) P-value

Age (Years) 0.88 (0.87 to 0.88) <0.001

AMTS 1.44 (1.34 to 1.54) <0.001

ASA 0.74 (0.66 to 0.84) <0.001

Mobility

Walk independently without aids

Walk with the aid of 1 stick

1.0 (ref)

0.32 (0.28 to 0.35)

-

<0.001

Sex

Female

Male

1.0 (Ref)

0.85 (0.77 to 0.93)

-

0.002

Fracture Volume (Deciles) 1.02 (0.97 to 1.08) 0.46

Day of Admission

Friday

Saturday

Sunday

Monday

Tuesday

Wednesday

Thursday

1.0 (ref)

1.01 (0.88 to 1.16)

1.03 (0.89 to 1.19)

1.09 (0.95 to 1.26)

1.11 (0.96 to 1.27)

1.17 (1.02 to 1.35)

1.17 (1.02 to 1.34)

-

0.92

0.66

0.21

0.16

0.03

0.03

Weekday Admission* 1.12 (1.04 to 1.21) 0.01

Deprivation Quintile

1 – Most deprived

2 -

3 -

4 -

5 – Least deprived

1.0 (ref)

1.07 (0.932to 1.24)

1.20 (1.04 to 1.38)

1.28 (1.11 to 1.47)

1.30 (1.13 to 1.51)

-

0.39

0.01

<0.001

<0.001

Date of surgery (6-months) 1.13 (1.10 to 1.15) <0.001

Page 10 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

* “Weekend” was defined as a Friday or Saturday admission, as surgery most

commonly occurs on the day following admission. ‘Weekday’ was therefore defined

as Sunday-Thursday. “Weekend” was included in the logistic model as a

dichotomous variable and day-of-the-week excluded as collinear.

Discussion

This observational study used a large national dataset and has demonstrated

unexplained variation in access to THA following hip fracture. Access to THA is

influenced by a number of patient characteristics, including age, sex, AMTS, ASA,

socioeconomic status, and pre-fracture mobility. Other key determinants were

the treating hospital and the day-of-the week of admission. Access to THA

amongst eligible patients improved over the study period but remains both low

and variable.

Compliance with NICE recommendations

NICE was established in 1999 to promote evidence-based treatments and reduce

unexplained variation in care across the NHS, the so-called “postcode lottery”.15

In June 2011, NICE recommended that THA should be offered to patients with a

displaced intracapsular hip fracture who can walk independently outdoors (with

no more than a single mobility aid), are cognitively intact, and are medically fit to

undergo the operation. This guideline is consistent with a developing evidence

base, which suggests that THA leads to better functional outcomes than HA

following hip fracture7 9-11. However, our study found that variation in access to

THA persists across the NHS because of poor compliance with NICE guidelines.

There was substantial variation in compliance (0.1 to 60.0%) between hospitals.

As patient-level predictors were unable to account for this variation, it is likely to

reflect systematic differences in practice between centres.

The optimal recursive partitioning model suggested that surgeons might

consider factors that could be relevant even if not strictly included within the

NICE guidelines. For example, older patients were less likely to undergo THA, as

were those that mobilised using a stick compared to those mobilising

Page 11 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyindependently without aids. Although there is strong evidence that some hip

fracture patients benefit from THA7 9-11, its precise indications are not well

defined. Our model offers a glimpse into the collective judgment of orthopaedic

surgeons and could be used to help inform the development of future NICE

guidelines in the absence of higher-level evidence. It is nevertheless concerning

that deprivation quintile was inversely associated with access to THA. There are

many potential explanations for this observation, including patient preferences

and confounding factors. However, it is also possible that heuristic judgments

about which patients are sufficiently “independent” to benefit from THA could be

influenced by implicit surgeon bias. Social class biases have been shown to

influence treatment decisions across a range of settings16-18 and could raise a

barrier for patients that are otherwise eligible to undergo THA. This inverse

association risks exacerbating health inequalities and is a further reason to

promote clear, evidence-based, national guidelines.

Barriers to increased THA provision

One potential obstacle to delivering THA for all eligible hip fracture patients is

the availability of experienced hip surgeons. It is widely accepted that patients

undergoing elective THA by a low-volume surgeon have greater risks of

dislocation, need for revision surgery, post-operative complications, and death19-

23. For this reason, many orthopaedic surgeons do not perform THA for hip

fracture if this operation is not part of their routine elective practice. The limited

availability of suitably experienced hip surgeons might account for the reduced

THA access observed at weekends. This finding is important in the context of

recent proposals to introduce seven-day services across the NHS24. Although this

discussion is principally framed around increased weekend mortality25 26, timely

access to THA for fracture may also need to be addressed. Regionalisation of hip

fracture services is one possible means of ensuring equal access to THA

throughout the week. High hospital case volume was not however associated

with improved access to THA.

Strengths and limitations of study

Page 12 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyThe main strength of this study was its use of a dataset that captures almost

every (>95%) hip fracture treated in England, Wales, and Northern Ireland.

There were variables that aligned closely with the NICE eligibility criteria, which

permitted the recommended treatment algorithm to be mapped over the

administrative data recorded within the NHFD.

The principal limitation was that the NHFD does not record individual patient

comorbidities. Some of the variables in our analysis (e.g. age and deprivation)

could simply represent a tendency towards a greater comorbidity burden.

However, ASA score has been shown to have equivalent or even greater

predictive value for mortality and complications than standard comorbidity

measures, such as the Charlson Comorbidity Index)27-29. It is unlikely that

patients assigned an ASA score <2 (2 = “mild systemic disease”) were medically

unfit to undergo THA. The NHFD also does not include sufficient detail to

understand clinical decision-making at the individual patient level. For example,

it is possible that THA was discussed with some patients and HA chosen

following a balanced risk-benefit discussion. However, the inter-hospital

variation in compliance with NICE guidelines suggests the existence of

systematic problems with THA provision.

Conclusion

Compliance with NICE guidance on THA for hip fracture is poor, with many

eligible patients failing to undergo THA. There continues to be substantial inter-

hospital variation in practice, which is not readily explained by patient-level

differences. The limited provision of THA to patients from deprived areas and

those requiring treatment at the weekend are particular concerns as it is

unacceptable for such decisions to depend on when and where a patient seeks

treatment for their hip fracture. There have been substantial improvements in all

of the quality indicators measured by the NHFD since its creation in 2004.3 The

NHFD should consider reporting data on THA provision at the hospital-level to

help achieve greater consistency across the NHS.

Page 13 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyCompeting interests

All authors have completed the ICMJE uniform disclosure form at

www.icmje.org/coi_disclosure.pdf and declare: no support from any

organisation for the submitted work; no financial relationships with any

organisations that might have an interest in the submitted work in the previous

three years; no other relationships or activities that could appear to have

influenced the submitted work.

Licence

The Corresponding Author has the right to grant on behalf of all authors and

does grant on behalf of all authors, a worldwide licence

(http://www.bmj.com/sites/default/files/BMJ%20Author%20Licence%20Marc

h%202013.doc) to the Publishers and its licensees in perpetuity, in all forms,

formats and media (whether known now or created in the future), to i) publish,

reproduce, distribute, display and store the Contribution, ii) translate the

Contribution into other languages, create adaptations, reprints, include within

collections and create summaries, extracts and/or, abstracts of the Contribution

and convert or allow conversion into any format including without limitation

audio, iii) create any other derivative work(s) based in whole or part on the on

the Contribution, iv) to exploit all subsidiary rights to exploit all subsidiary rights

that currently exist or as may exist in the future in the Contribution, v) the

inclusion of electronic links from the Contribution to third party material where-

ever it may be located; and, vi) licence any third party to do any or all of the

above. All research articles will be made available on an Open Access basis (with

authors being asked to pay an open access fee—see

http://www.bmj.com/about-bmj/resources-authors/forms-policies-and-

checklists/copyright-open-access-and-permission-reuse). The terms of such

Open Access shall be governed by a Creative Commons licence—details as to

which Creative Commons licence will apply to the research article are set out in

our worldwide licence referred to above.

Page 14 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyTransparency Declaration

Daniel Perry (the manuscript’s guarantor) affirms that the manuscript is an

honest, accurate, and transparent account of the study being reported; that no

important aspects of the study have been omitted; and that any discrepancies

from the study as planned (and, if relevant, registered) have been explained.

What this paper adds

1. What is already known on this subject

A defined subset of hip fracture patients achieve better functional outcomes from

total hip arthroplasty (THA) than hemiarthroplasty. Guidelines published by the

National Institute for Health and Care Excellence (NICE) clearly indicate which

hip fracture patients should be offered THA.

2. What this study adds

Compliance with NICE guidelines is poor and there is considerable variation

between hospitals. Surgeons appear to apply different eligibility criteria for THA

than NICE. Socioeconomic deprivation and need for hip fracture surgery at the

weekend are particular barriers to accessing THA. Further efforts are necessary

to improve access to THA for eligible patients and reduce unexplained variation

in care for older adults with hip fractures.

Page 15 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyFigure legends

Figure 1 – Decision tree for THA in displaced intracapsular fractures as per NICE

guidelines. Parentheses indicate the number of people within each leaf and the

percentage of people within the leaf that underwent THA.

Figure 2 – Graph illustrating limited improvement in the model using the optimal

tree size of ten terminal nodes (lowest error), and a more pragmatic tree with

five nodes. cp = complexity function.

Figure 3 – Decision tree for THA in displaced intracapsular fractures using a

recursive partitioning algorithm. Parentheses indicate the number of people

within each leaf and the percentage of people within the leaf that underwent

THA.

Figure 4 – Graph illustrating limited improvement in the model using the optimal

tree size of 20 terminal nodes, and a simplified tree with four nodes.

Figure 5 - Decision tree using a recursive partitioning algorithm to indicate the

important predictors for THA amongst individuals fulfilling the NICE criteria for

consideration of THA. Parentheses indicate the number of people within each

leaf and the percentage of people within the leaf that underwent THA.

Figure 6 – Variation in the number of THAs performed within each hospital, as a

proportion of the total number of individuals fulfilling the NICE guidelines. Only

hospitals that contributed >100 NICE eligible patients are included to minimise

spurious data (n=96). Each bar represents a hospital.

Page 16 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyReferences

1. Royal College of Physicians. National Hip Fracture Database annual report

2014. London, UK: RCP 2014.

2. Burge RT, Worley D, Johansen A, et al. The cost of osteoporotic fractures in the

UK: projections for 2000-2020. J Med Econ 2001;4:51-62.

3. Neuburger J, Currie C, Wakeman R, et al. The impact of a national clinician-led

auditinitiative on care and mortality after hip fracture in England: an

external evaluation using time trends in ton-audit tata. Med Care

2015;53(8):686-91.

4. Sahota O, Currie C. Hip fracture care: all change. Age Ageing 2008;37(2):128-9.

5. British Orthopaedic Association, British Geriatrics Society. The Care of Patients

with Fragility Fracture. London: BOA 2007.

6. Department of Health. Payment by Results Guidance for 2013-14. London:

DoH 2013.

7. Hopley C, Stengel D, Ekkernkamp A, et al. Primary total hip arthroplasty versus

hemiarthroplasty for displaced intracapsular hip fractures in older

patients: systematic review. BMJ 2010;340:c2332.

8. National Institute for Health and Care Excellence. Hip fracture: the

management of hip fracture in adults. NICE clinical guideline 124. London,

NICE 2011.

9. Parker MJ, Gurusamy KS, Azegami S. Arthroplasties (with and without bone

cement) for proximal femoral fractures in adults. The Cochrane Database

of Systematic Reviews 2010(6):CD001706.

10. Avery PP, Baker RP, Walton MJ, et al. Total hip replacement and

hemiarthroplasty in mobile, independent patients with a displaced

intracapsular fracture of the femoral neck: a seven- to ten-year follow-up

report of a prospective randomised controlled trial. J Bone Joint Surg Br

2011;93(8):1045-8.

11. Yu L, Wang Y, Chen J. Total hip arthroplasty versus hemiarthroplasty for

displaced femoral neck fractures: meta-analysis of randomized trials. Clin

Orthop Relat Res 2012;470(8):2235-43.

12. Department of Health. Governance arrangements for research ethics

committees. London: DoH 2011.

13. McLennan D, Barnes H, Noble M, et al. The English Indices of Deprivation

2010. London: Department for Communities and Local Government 2011.

14. Therneau TM, Atkinson EJ. An introduction to recursive partitioning using

rpart routines: The Comprehensive R Archive Network (CRAN), 2015.

15. Vyawahare B, Hallas N, Brookes M, et al. Impact of the National Institute for

Health and Care Excellence (NICE) guidance on medical technology

uptake: analysis of the uptake of spinal cord stimulation in England 2008-

2012. BMJ Open 2014;4(1):e004182.

16. Haider AH, Schneider EB, Sriram N, et al. Unconscious race and social class

bias among acute care surgical clinicians and clinical treatment decisions.

JAMA Surg 2015;150(5):457-64.

17. Tamayo-Sarver JH, Dawson NV, Hinze SW, et al. The effect of race/ethnicity

and desirable social characteristics on physicians' decisions to prescribe

opioid analgesics. Acad Emerg Med 2003;10(11):1239-48.

Page 17 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly18. Street RL, Jr., O'Malley KJ, Cooper LA, et al. Understanding concordance in

patient-physician relationships: personal and ethnic dimensions of shared

identity. Ann Fam Med 2008;6(3):198-205.

19. Hedlundh U, Ahnfelt L, Hybbinette CH, et al. Surgical experience related to

dislocations after total hip arthroplasty. J Bone Joint Surg Br

1996;78(2):206-9.

20. Lavernia CJ, Guzman JF. Relationship of surgical volume to short-term

mortality, morbidity, and hospital charges in arthroplasty. J Arthroplasty

1995;10(2):133-40.

21. Katz JN, Losina E, Barrett J, et al. Association between hospital and surgeon

procedure volume and outcomes of total hip replacement in the United

States medicare population. J Bone Joint Surg Am 2001;83-A(11):1622-9.

22. Losina E, Barrett J, Mahomed NN, et al. Early failures of total hip replacement:

effect of surgeon volume. Arthritis Rheum 2004;50(4):1338-43.

23. Katz JN, Phillips CB, Baron JA, et al. Association of hospital and surgeon

volume of total hip replacement with functional status and satisfaction

three years following surgery. Arthritis Rheum 2003;48(2):560-8.

24. Kleebauer A, Comerford C. Government commits to seven-day NHS. Nurs

Manage 2015;22(3):6.

25. Freemantle N, Richardson M, Wood J, et al. Weekend hospitalization and

additional risk of death: an analysis of inpatient data. J R Soc Med

2012;105(2):74-84.

26. Keogh B. Should the NHS work at weekends as it does in the week? Yes. BMJ

2013;346:f621.

27. Whitmore RG, Stephen JH, Vernick C, et al. ASA grade and Charlson

Comorbidity Index of spinal surgery patients: correlation with

complications and societal costs. Spine J 2014;14(1):31-8.

28. Tan WP, Talbott VA, Leong QQ, et al. American Society of Anesthesiologists

class and Charlson's comorbidity index as predictors of postoperative

colorectal anastomotic leak: a single-institution experience. J Surg Res

2013;184(1):115-9.

29. Dekker JW, Gooiker GA, van der Geest LG, et al. Use of different comorbidity

scores for risk-adjustment in the evaluation of quality of colorectal cancer

surgery: does it matter? Eur J Surg Oncol 2012;38(11):1071-8.

Page 18 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

Figure 1

240x128mm (72 x 72 DPI)

Page 19 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

Figure 2

278x153mm (72 x 72 DPI)

Page 20 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

Figure 3

240x128mm (72 x 72 DPI)

Page 21 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

Figure 4

228x125mm (72 x 72 DPI)

Page 22 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

Figure 5

240x128mm (72 x 72 DPI)

Page 23 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

Figure 6

289x420mm (72 x 72 DPI)

Page 24 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyMobility Scores

Version 7 and 8 of NHFD have different mobility scores

V7 therefore tracked onto V8

MOBILITY (VERSION 7 ONLY)

WalkInside Walking Ability Indoors (V7 only)

0 Regulary walked without aids

1 Regularly walked with one aid

2 Regularly walked with two aids or frame

3 Wheelchair or bedbound

4 Unknown

. Missing

WalkOutside Walking ability outside (V7 only)

0 Regulary walked without aids

1 Regularly walked with one aid

2 Regularly walked with two aids or frame


4 Never goes outdoors

5 Electric buggy

6 Unknown

. Missing

AccompOutside Accompanied to Walk Ouside (V7 only)

0 No

1 Yes


3 Unknown

4 Never goes outdoors

. Missing

AccompInside Accompanied to Walk Inside (V7 only)

0 No

1 Yes


3 Unknown

. Missing

MOBILITY VERSION 8 ONLY

Mobility Pre-Fracture Walking Ability (VERSION 8 ONLY)

1 Freely mobile without aids

2 Mobile outdoors with one aid

3 Mobile outdoors with two aids or frame

Page 25 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly4 Some indoor mobility but never goes outside without help

5 Unknown

6 No functional mobility

V7 TRACKED ONTO V8 (variable name = mob)

1 Freely mobile without aids (walkoutside 0) 2 Mobile outdoors with one aid (walkoutside 1) 3 Mobile outdoors with two aids or frame (walkoutside 2)

4

Some indoor mobility but never goes outside without help

ANY WALKING INSIDE(0-2) AND ACCOMPANIEDOUTSIDE (1,2 and 4) OR ANYWALKING INSIDE(0-2) AND WALKINGOUTDOORS WHEEL/ BUGGY (3, 4,5)

5 No functional ability Wheelchair or bedbound inside (walkinside 3) 6 Unknown

STATA CODE

replace mob=mobility

replace mob=5 if walkinside==3

replace mob=4 if (walkinside==0 & accompoutside==1)









replace mob=4 if (walkinside==0 & walkoutside==3)




Page 26 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyreplace mob=4 if (walkinside==1 & walkoutside==4)





replace mob=3 if walkoutside==2



Page 27 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyGroup ‘X’ Hospitals

ADD. Addenbrookes Hospital

AEI. Royal Albert Edward Infirmary

AIR. Airedale General Hospital

ASH. Wansbeck General Hospital

BAR. Barnsley District General Hospital

BAS. Basildon Hospital

BAT. Royal United Hospital Bath

BFH. Broomfield Chelmsford

BOL. Royal Bolton Hospital

BRD. Bradford Royal Infirmary

BRG. Bronglais General Hospital

BRI. Bristol Royal Infirmary

CCH. Charing Cross Hospital

CHE. Chesterfield Royal

CHG. Cheltenham General Hospital

COC. Countess of Chester Hospital

CRG. Craigavon Area Hospital

DAR. Darlington Memorial Hospital

DGE. Eastbourne DGH

DVH. Darent Valley Hospital

EAL. Ealing Hospital

EBH. Birmingham Heartlands Hospital

ESU. East Surrey Hospital

FAZ. University Hospital Aintree

FGH. Furness General

FRM. Frimley Park Hospital

FRY. Frenchay Hospital

GGH. Diana Princess of Wales Hospital

GHS. Good Hope General Hospital

GLO. Gloucestershire Royal Hospital

GRA. Grantham And District General Hospital

GWY. Ysbyty Gwynedd Hospital

HAR. Harrogate District Hospital

HCH. County Hospital Hereford

HIN. Hinchingbrooke Hospital

HOM. Homerton Hospital

HRI. Hull Royal Infirmary

IOW. St Marys Hospital Newport

JPH. James Paget Hospital

LDH. Luton & Dunstable Hospital

LGH. Leighton Hospital

LGI. Leeds General Infirmary

LIN. Lincoln County Hospital

LON. Royal London Hospital

Page 28 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyMAC. Macclesfield District General Hospital

MKH. Milton Keynes General Hospital

MOR. Morriston Hospital

MPH. Taunton & Somerset Hospital

NCR. New Cross Hospital

NEV. Nevill Hall Hospital

NGS. Northern General Hospital

NHH. North Hampshire Hospital

NMG. North Manchester General Hospital

NMH. North Middlesex Hospital

NTG. University Hospital of North Tees

NTH. Northampton General Hospital

NUH. Ulster Hospital

OHM. Royal Oldham Hospital

PAH. Princess Alexandra Hospital

PIN. Pinderfields General Hospital

PLY. Derriford Hospital

PMS. The Great Western Hospital

QAP. Queen Alexandra Hospital

QEB. Queen Elizabeth Hospital Edgbaston

QEG. Queen Elizabeth Hospital Gateshead

QKL. Queen Elizabeth Hospital (Kings Lynn)

RAD. John Radcliffe Hospital

RBE. Royal Berkshire Hospital

RCH. Royal Cornwall Hospital

RDE. Royal Devon & Exeter Hospital

RED. The Alexandra Hospital

RFH. Royal Free Hospital

RGH. Royal Glamorgan

RHC. Royal Hampshire County Hospital

RLI. Royal Lancaster Infirmary

RLU. Royal Liverpool University Hospital

ROT. Rotherham General Hospital

RPH. Royal Preston Hospital

RSC. Royal Sussex County Hospital

RSS. Royal Shrewsbury Hospital

RSU. Royal Surrey County Hospital

RVB. Royal Victoria Hospital

RVN. Royal Victoria Infirmary

SAL. Salisbury District Hospital

SAN. Sandwell District Hospital

SCA. Scarborough General Hospital

SCM. James Cook University Hospital

SDG. Staffordshire General Hospital

SEH. Southend Hospital

SGH. Southampton General Hospital

Page 29 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlySHH. Stepping Hill Hospital

SLF. Hope Hospital

SMV. Stoke Mandeville Hospital

SOU. Southport and Formby District General

SPH. St Peters Hospital

STD. South Tyneside District Hospital

STH. St Thomas Hospital

STM. St Marys Hospital Paddington

STO. University Hospital of North Staffordshire

STR. St Richards Hospital

SUN. Sunderland Royal Hospital

TOR. Torbay Hospital

TRA. Trafford General Hospital

TUN. Tunbridge Wells Hospital

UCL. University College Hospital

UHC. University Hospital Coventry

UHW. University Hospital of Wales

WAR. Warwick Hospital

WDG. Warrington District General Hospital

WES. Chelsea & Westminister Hospital

WEX. Wexham Park Hospital

WGH. Weston General Hospital

WHC. Whipps Cross Hospital

WHH. William Harvey Hospital

WHI. Whiston Hospital

WIR. Arrowe Park Hospital

WRG. Worthing Hospital

WRX. Maelor Hospital

WSH. West Suffolk Hospital

WWG. West Wales General

YDH. York District Hospital

YEO. Yeovil District Hospital

ENH. East and North Herts Hospital

QEW. Queen Elizabeth II Hospital (Welwyn)

Page 30 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyGroup ‘Y’ Hospitals

ALT. Altnagelvin Hospital

BED. Bedford Hospital

BLA. Royal Blackburn Hospital

BNT. Barnet General Hospital

BRO. Princess Royal University Hospital

(Bromley)

BRT. Queens Hospital

BRY. Fairfield General Hospital

BSL. Bassetlaw District General Hospital

CHS. Chase Farm Hospital

CLW. Glan Clwyd DGH Trust

CMI. Cumberland Infirmary

COL. Colchester General Hospital

DER. Royal Derby Hospital

DID. Doncaster Royal Infirmary

DRY. University Hospital of North Durham

GEO. St Georges Hospital

GWE. Royal Gwent Hospital

GWH. Queen Elizabeth Hospital Woolwich

HIL. Hillingdon Hospital

HOR. Horton General Hospital

HUD. Huddersfield Royal Infirmary

IPS. The Ipswich Hospital

KCH. Kings College Hospital

KGH. Kettering General Hospital

KMH. Kings Mill Hospital

KTH. Kingston Hospital

LER. Leicester Royal Infirmary

LEW. University Hospital Lewisham

MAY. Mayday University Hospital

MDW. Medway Maritime Hospital

MRI. Manchester Royal Infirmary

NDD. North Devon District Hospital

NOB. Nobles Hospital

NOR. Norfolk and Norwich Hospital

NPH. Northwick Park Hospital

NTY. North Tyneside Hospital

NUN. George Eliot Hospital

NWG. Newham General Hospital

OLD. Queens Hospital Romford

PCH. Prince Charles Hospital

PET. Peterborough City Hospital

PGH. Poole General Hospital

PIL. Pilgrim Hospital

Page 31 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlyPOW. Princess Of Wales Hospital

QEQ. Queen Elizabeth the Queen Mother

Hospital

RUS. Russells Hall Hospital

SCU. Scunthorpe General Hospital

SHC. St Helier Hospital

TGA. Tameside General Hospital

TLF. Princess Royal Hospital Telford

UHN. University Hospital Queens Medical

Centre

VIC. Victoria Hospital

WAT. Watford General Hospital

WDH. Dorset County Hospital

WHT. Whittington Hospital

WMH. Manor Hospital

WMU. West Middlesex University Hospital

WRC. Worcestershire Royal Hospital

WYB. Withybush General Hospital

WYT. Wythenshawe Hospital

MAI. Maidstone General Hospital

Hospitals not contributing to analysis owing to absent

or insufficient data

KSX. Kent & Sussex Hospital

CGH. Conquest Hospital

CMH. Central Middlesex Hospital

DMO. Demonstration Medical Centre

PEH. Princess Elizabeth Hospital

SHJ. Jersey General Hospital

WCI. West Cumberland Infirmary

LIS. Lister Hospital

Page 32 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only

Inequalities in Access to Total Hip Arthroplasty for Hip Fracture: Population Based Study

Item

No Recommendation

Title and abstract

Y (a) Indicate the study's design with a commonly used term in the title or the abstract

(b) Provide in the abstract an informative and balanced summary of what was done and what was found

Introduction

Background/rationale Y Explain the scientific background and rationale for the investigation being reported

Objectives Y State specific objectives, including any prespecified hypotheses

Methods

Study design Y Present key elements of study design early in the paper

Setting Y Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data

collection

Participants Y

(a) Cohort study?Give the eligibility criteria, and the sources and methods of selection of participants. Describe

methods of follow-upCase-control study?Give the eligibility criteria, and the sources and methods of case

ascertainment and control selection. Give the rationale for the choice of cases and controlsCross sectional study?Give

the eligibility criteria, and the sources and methods of selection of participants

(b) Cohort study?For matched studies, give matching criteria and number of exposed and unexposedCase-control

study?For matched studies, give matching criteria and the number of controls per case

Variables Y Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic criteria,

if applicable

Data sources/ measurement Y For each variable of interest, give sources of data and details of methods of assessment (measurement). Describe

comparability of assessment methods if there is more than one group

Bias Y Describe any efforts to address potential sources of bias

Study size Y Explain how the study size was arrived at

Quantitative variables Y Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen

and why

Statistical methods Y

(a) Describe all statistical methods, including those used to control for confounding

(b) Describe any methods used to examine subgroups and interactions

(c) Explain how missing data were addressed

(d) Cohort study?If applicable, explain how loss to follow-up was addressedCase-control study?If applicable, explain

how matching of cases and controls was addressedCross sectional study?If applicable, describe analytical methods

Page 33 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only

taking account of sampling strategy

(e) Describe any sensitivity analyses

Results

Participants Y

(a) Report numbers of individuals at each stage of study?eg numbers potentially eligible, examined for eligibility,

confirmed eligible, included in the study, completing follow-up, and analysed

(b) Give reasons for non-participation at each stage

(c) Consider use of a flow diagram

Descriptive data Y

(a)Give characteristics of study participants (eg demographic, clinical, social) and information on exposures and

potential confounders

(b) Indicate number of participants with missing data for each variable of interest

(c) Cohort study?Summarise follow-up time (eg average and total amount)

Outcome data Y

Cohort study?Report numbers of outcome events or summary measures over time

Case-control study?Report numbers in each exposure category, or summary measures of exposure

Cross sectional study?Report numbers of outcome events or summary measures

Main results Y

(a) Report the numbers of individuals at each stage of the study?eg numbers potentially eligible, examined for

eligibility, confirmed eligible, included in the study, completing follow-up, and analysed

(b) Give reasons for non-participation at each stage

(c) Consider use of a flow diagram

Other analyses Y Report other analyses done?eg analyses of subgroups and interactions, and sensitivity analyses

Discussion

Key results Y Summarise key results with reference to study objectives

Limitations Y Discuss limitations of the study, taking into account sources of potential bias or imprecision. Discuss both direction and

magnitude of any potential bias

Interpretation Y Give a cautious overall interpretation of results considering objectives, limitations, multiplicity of analyses, results from

similar studies, and other relevant evidence

Generalisability Y Discuss the generalisability (external validity) of the study results

Other information

Funding Y Give the source of funding and the role of the funders for the present study and, if applicable, for the original study on

which the present article is based

Page 34 of 34


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Documents

Inequalities in Access to Total Hip Arthroplasty for Hip Fracture: … · 2016-04-29 · Confidential: For Review Only Inequalities in Access to Total Hip Arthroplasty for Hip Fracture: