Академический Документы
Профессиональный Документы
Культура Документы
ABSTRACT RESUME
Background: The United Kingdom (UK) National Health Service Contexte : Le Rapport sur les activites d’imagerie et de radiodiagnos-
(NHS) Imaging and Radiodiagnostic activity 2013/14 report esti- tic 2013-2014 du National Health Service (NHS) du Royaume-Uni
mates the year-on-year increase of magnetic resonance imaging estime que l’augmentation annuelle des examens d’imagerie par
(MRI) examinations to be 12.3%, with the designated radiologist resonance magnetique (IRM) a ete de 12,3 %, avec une dispropor-
workforce disproportionate to the increase in demand. tion entre le nombre de radiologistes designes et l’augmentation de
la demande en rapports d’imagerie.
Objective: To review the economics, risk, and feasibility of MRI re-
porting by radiographers. Objectif : Examiner l’aspect economique, le risque et la faisabilite de
faire preparer les rapports d’IRM par les radiographes.
Design: A PICO (the four major components of a clinical or
research question: patient [population], intervention, comparison, Conception : Un cadre PICR (les quatre elements d’une question
and outcome) framework using example patient demand from clinique ou d’une recherche: patient (population), intervention, com-
audit data of noncomplex MRI examination attendance (n ¼ paraison et resultat) utilisant la demande des patients tiree des
3,525) over 12 months was used to review costs, potential outcome donnees de verification de presence a un examen d’IRM non com-
risks (diagnostic thresholds), and feasibility (workforce capacity) of plexe (n ¼ 3 525) sur une periode de 12 mois, avec un examen
both interventions. des co^uts, des risques potentiels pour les resultats (seuil de diagnostic)
et de la faisabilite (capacite des effectifs) pour les deux interventions.
Conclusions: The benefits of introducing a skills mix reporting ser-
vice model to the benefit of service delivery in the UK has shown a Conclusions : Les avantages liesa l’introduction d’un service de rapport
potential £145,230–£60,524 per annum cost saving using a generic en competences mixtes pour la prestation des services au Royaume-Uni
acute workload model. Research into recorded discrepancy/error montrent des economies annuelles potentielles de l’ordre de £145 230 -
audit data for potential detrimental risk to patient outcomes identi- £60 524 par annee avec l’utilisation d’un modele generique de charge de
fied a paucity of evidence and recommends that further research is travail aigu€e. Les recherches sur les donnees de verifications des ecarts ou
needed. erreurs enregistrees afin d’etablir le risque potentiel pour les patients ont
permis de constater en manque de donnees probantes, et il est re-
commande de poursuivre les recherches sur cet aspect.
Keywords: Magnetic resonance imaging; reporting radiographer; role extension; advanced practice; economic evaluation
1939-8654/$ - see front matter Ó 2016 Canadian Association of Medical Radiation Technologists. Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jmir.2015.12.083
by improving value rather than reducing the costs, in the wake The study used data from a retrospective audit of MRI exam-
of the Five Year Forward View [1] that proposed £22 billion ination attendance at an acute NHS district general hospital
of efficiency savings. The report findings emphasised the need (DGH). A defined time horizon of 12 months (August
to create an environment for change and highlighted major 2014–July 2015; Tables 1 and 2) was used to identify the
service areas where restructuring diagnostic pathways could key resource demand for MRI examinations (n ¼ 12,958).
deliver cost-effective service improvements, increase the speed Using decision tree modelling to illustrate the process
of delivery, reduce length of stay in hospital, and fast-track mapping of the current intervention (Table 3) allowed evalu-
treatment and management for preventable illness. ation of costs and outcomes from each intervention for inter-
Two specific driving factors for change in radiology have nal validity. Using the audit data allowed external validation
been a flexible response to workforce shortages [5–10], and of the model as an example of expected workflow demand
demand for imaging that outstrips capacity [11–13]. The in a generic DGH. A decision tree was chosen over conven-
NHS Imaging and Radiodiagnostic activity 2013/14 report tional Markov models because data for chronic returning pa-
[11] assessed the number of magnetic resonance imaging tients were not available to consider all feasible transitions of
(MRI) examinations from April 2013 to March 2014 to be patient’s health states or cohorts of particular disease categor-
2.7 million, with a 12.3% increase in examinations from ised patients.
the previous year [11], 71.7% over 5 years [14], and 220%
Patient Group
growth over a 10-year period [11], which is a substantial in-
crease in the pattern and trend of imaging demand. Both The sample size from the data collection identified n ¼
the RCR [13] and the Society and College of Radiographers 3,525 noncomplex MRI scans (Table 2). The inclusion
[13] (SCoR) observe that demand in imaging is expected to criteria included knee, lumbar, internal auditory meatus
intensify. The Centre for Workforce Intelligence (CfWI) (IAMs), scaphoid, and breast. The noncomplex examination
[12] predicts the demand for imaging will escalate due to criteria limitations were due to the restricted literature evi-
multiple factors, including growing and/or aging populations, dence available on reported diagnostic thresholds of reporting
rise in cancer diagnosis and chronic illness, screening pro- radiographers and radiologists in MRI reporting.
grammes, introduction of 24/7 working hours, and future im-
The Current Intervention
aging techniques introduced into clinical practice. The CfWI
and the RCR have estimated the total imaging workload The NHS currently employs radiologists to report MRI
could potentially rise from 39 million tests in 2011 to 51 examinations, but drivers for change include the low work-
million by 2025 (an increase of all imaging by 76%, with force numbers of UK registered radiologists [14]. The fifth
MRI alone raising 87%) [12]. RCR workforce report 2012 [14] recorded the number of
The RCR [15] recommend a formal report for diagnostic UK registered radiologists as 2,997 (4.7 radiologists per
examinations within 2 days, but acknowledge that due to 100,000 population); with a current deficit of 421 vacant
workforce shortages, this is not occurring [14]. This causes de- posts [3]. To reach comparable radiologist levels with the
lays in cancer and serious illness diagnosis, hospital stay, and rest of the European Union countries, the RCR estimated it
the subsequent increased listing of radiology departments to would require an 82% increase of radiologists [12].
NHS risk registers [15]. In October 2014, an RCR survey The CfWI report on Clinical Radiology [12] commis-
[16] highlighted a month delay in results (1,697 examina- sioned by the Department of Health (DoH) with multiple
tions) in the 25% of NHS trusts surveyed. The survey was stakeholders, including the RCR and SCoR, reviewed the
repeated in February 2015 [15] with 71% of surveyed trusts RCR 2012 [13] report for the Medical Programme Board
having delays of more than a month, with over 3,277 unre- and the Joint Working Group on Speciality Training
ported MRI scans (estimated for all trusts in England to be Numbers. Recommendations included a proposed, but not
up to 4,268 [15]). implemented, increase of 60 trainee radiology registrars per
year, with the use of radiographers to effectively support the
Methodology
future expansion of radiology.
To define the perspective of the review and the key drivers of
Unit Costs and Discounting
cost-effectiveness (capacity and demand, benefits, and risks) a
PICO framework [17] was adopted. A PICO has four major To ascertain an average hourly price for radiologists, Net-
components, in this case consisting of P for the patient sample ten et al’s Ready Reckoner for staff costs in the NHS [18] and
group defined by the MRI imaging pathway; I for the intervention the Personal Social Services Research Unit (PSSRU) Unit
of radiographers reporting MRI examinations; C for the compar- Costs of Health and Social Care 2014 [19] were adopted
ison to existing intervention of radiologists reporting MRI for the basis of the calculations. The salary was based on a
examinations; and O for the outcome comparison of current full time equivalent (FTE) mean of NHS medical consultant
and alternative service provision through costs, savings, and risk. wages [19]. An additional 33.5% was added for overtime,
This review received university research ethical and gover- shift work and geographic allowances [19], National Insur-
nance approval to calculate a deterministic, scenario-based ance contributions (NI) [20], and employer’s contribution
evaluation of costs for the current and new intervention. to superannuation [21]. The costs for education and training
used PSSRU [19] standard estimation approaches for the and radiologists working in complementary reporting roles
components of training, tuition fees, clinical placement costs, (not substitution or replacement of roles) to sustain service de-
infrastructure (books, journals, and computers), and lost pro- livery. The SCoR scope of practice [27,28] legally entitles UK
duction costs of staff training days. radiographers with accredited training and competence to
The costs incorporated the discounting system used by report MRI examinations. The CfWI [29] have predicted
PSSRU [19] and Her Majesty’s Treasury [22] to transfer all an increase of 17% (to 19,830) of radiographers from 2012
costs and benefits to ‘‘present values’’ to compare, using a to 2016. Currently, the Health and Care Professions Council
3.5% discount rate. This allowed a net present value of the (HCPC) [30] have 29,711 radiographers registered (3,100 are
intervention to be calculated, which is the primary indicator therapeutic radiographers [31]), which is above the projected
used by the UK government to justify action. Furthermore, increase of workforce by the CfWI [29]. In addition, Health
this is the approved system in use by the National Institute Education England (HEE) [32] has increased educational
for Health and Care Excellence (NICE) [23] for all DoH commissioning of places for 2014/2015.
[24] assessment and appraisals of health technologies, tech- The 2014 UK radiographer unfilled vacancy rate was 5.1%
niques, and screening programmes. The hourly unit cost of at Band 7 reporting level [31]; the SCoR [31] estimate 3,662
a radiologist (2014–2015) was calculated at £156 (Table 4). radiographers were in advance practice and 86 in consultant
roles, with a further 1,288 in postgraduate training [31]. The
The New Intervention master’s degree pathway in clinical reporting in our university
The RCR and the SCoR have jointly published guidance currently offers a wide range of options for MRI reporting
[26] to endorse the collaborative skills mix of radiographers modulesdincluding head and neck, IAM, spine, breast,
Table 2
Audit Results of MRI Demand at an Average-Sized Generic DGH (2014–2015)
Area Aug-14 Sep-14 Oct-14 Nov-14 Dec-14 Jan-15 Feb-15 Mar-15 Apr-15 May-15 Jun-15 Jul-15 Total
Knee 166 110 122 127 125 152 102 150 106 105 116 120 1,501
Lumbar 127 102 115 93 149 103 122 109 88 106 102 104 1,320
IAM 33 39 39 41 34 41 47 47 45 33 38 25 462
Scaphoid 17 11 15 17 14 10 7 7 21 16 13 13 161
Breast 2 9 5 8 5 7 5 6 9 4 11 10 81
All MRI 1,189 1,093 1,136 1,057 1,051 1,109 1,064 1,180 1,016 971 1,035 1,057 12,958
DGH, district general hospital; IAM, internal auditory meatus; MRI, magnetic resonance imaging.
gastrointestinal, knee, foot and ankledwith strong recruitment rate was applied, and the hourly unit cost of a reporting ra-
of students. diographer (2014–2015) was determined as £53 (Table 5).
Table 4
Consultant Radiologist Hourly Unit Cost Calculation
Unit Costs and Unit Estimation 2014/2015 Value Notes
A Wages/salary (þ) £87,060 per year Medical consultant average [19]
£29,165 per year 33.5% Allowances [19] for overtime/shift work/etc
B Salary oncosts (þ) £5,012 per year National Insurance secondary threshold deduction [20]
(þ) £11,753 per year Superannuation–NHS pensions 13.5%–Tier 6 [21]
London multiplier 1.19 (AþB) and 1.39 G Allow for higher costs of living in London [19]
Non-London multiplier 0.97 (AþB) and 0.97 G Allow for lower costs of living outside London [19]
C Qualifications (þ) £72,197 per year Taken from PSSRU [19], using Netten et al [18] costs from
DoH and HEE consultants ¼ 2 foundation years, 6 speciality registrar years
D Fees (þ) £420 per year GMC [25]
E Overheads, management, Taken from PSSRU–NHS (England) [19]
Administration and estates staff (þ) £20,048 per year Management and noncare staff 19.31% of direct care salary
F Nonstaff (þ) £43,575 per year Nonstaff costs 41.97% of direct salary costs (include costs to provider–office,
travel/transport, telephone, education, training, supplies, services, utilities of
water, gas, electricity) [19]
G Capital overheads (þ) £8,411 per year Capital costs annuitised over 60 years (discount rate of 3.5%) based on
PSSRU [19]. New build and land requirements of NHS hospitals (adjusted
for both treatment and nontreatment space)
H Working time (O) 42.4 wk/y PSSRU [19] calculated unit costs of 1,589 h/y: 212
(O) 37.5 h/wk Working days (minus sickness absence and training) [19]
Unit costs 2014/2015 £156 per hour
DoH, Department of Health; GMC, General Medical Council; HEE, Health Education England; NHS, National Health Service; PSSRU, Personal Social
Services Research Unit.
10 minutes per examination per report, respectively [12]). of life) from the change of service delivery. The DGH audit
The CfWI and DoH [12] use weighted factors of 24, 16, data did not provide statistics from error and/or discrepancy
and 12 minutes per examination per report. The CfWI [12] meetings to assess the potential for detrimental risk to pa-
calculated that each FTE radiologist was allocated 10.3 pro- tient outcomes through reporting. To estimate the potential
grammed activities (PAs); 2 PAs for nonreporting administra- for errors in reporting that theoretically could affect the
tion of paperwork, teaching, and other duties, with 8 weeks health of patients, a literature review on the diagnostic accu-
deducted for annual leave/study. Likewise, the RCR [14] cal- racy of radiologists and reporting radiographers interpreting
culations use 10.3 PAs (8 PAs over 44 weeks). MRI examinations was performed using Cochrane, MED-
A limitation of this review found the SCoR have no pub- LINE, Europe PubMed Central, CINAHL, Science Direct,
lished costings of reporting radiographers’ unit costs per and Google Scholar. The search results were limited because
noncomplex MRI examination to compare against, so the of the variation and quality of the literature methodologies
RCR [12,36] and CfWI and DoH [12] systems have been and results, and details on sample size and pathology range
adopted for comparison. A literature search using CINAHL, were inconsistent. In addition, reference standards varied,
Cochrane Library, MEDLINE, PubMed, Science Direct, with certain studies only providing agreement levels, mostly
and Google Scholar found no studies on the time taken for without confidence intervals, sensitivity, or specificity.
radiographers to report MRI scans. The study reverted to Observer variation studies on radiologist diagnostic perfor-
the evidence of previously published studies [3739] that mance from a number of published sources comparing set
applied timed reporting of MRI examinations (same case- reference standards for MRI knee studies have identified radi-
loads) of radiographers and radiologists, which produced ologist agreement levels ranging from 48.1% to 96%
near equivalent accuracy, agreement, sensitivity, and speci- [37,4046]. Radiologist sensitivity levels ranged from
ficity results. 73.5% to 88% [42,43,46,47], with specificity between
90.8% to 97% [40,43,46,47]. The introduction of reporting
Comparison of Diagnostic Thresholds per Intervention radiographers to interpreting MRI knee examinations was re-
The risk of discordance in reporting is another important viewed by the university in an academic setting [39],
measure to include in the evaluation of assessing interven- recording a mean sensitivity of 99.4% (95% CI, 97.4–99.8)
tions. This will influence potential impact on patient out- [39] and mean specificity of 95.9% (95% CI, 93.1–97.7)
comes (mortality, morbidity, functional status, and quality [39].
Specificity %
ranged from 83.3% to 100% [5153], specificity 90%
% Unknown
% Unknown
100 [52,53]
to 100% [5153], and agreement of 86.65% to 100%
69.2 [56]
96.3 [54]
94 [39]
99 [38]
99 [38]
99 [38]
[5153]. Radiographers demonstrated a mean sensitivity
rate of 100% (95% CI, 82.3–95.1) [54] and specificity
96.3% (95% CI, 90.1–100) [54] and agreement of 92.2%
Sensitivity %
% Unknown
% Unknown
sensitivity and specificity levels of 99% [38].
100 [52,53]
88.6 [56]
95.2 [39]
100 [54]
IAM diagnostic threshold studies have identified radiolo-
99 [38]
99 [38]
99 [38]
gist agreement levels between 56% and 100% [37,55], with
no found levels of separate sensitivity or specificity. Reporting
radiographers agreement levels for IAMs were 98.4% [38],
88.6 [39]
92.2 [54]
98.4 [38]
100 [51]
85 [56]
Lumbar spine
Lumbar spine
Knee
IAM
IAM
Applying the CfWI and DoH [12] ranges to the hourly re-
porting rates estimated a cost difference of £41.20–£20.60
Radiographer–reporting
Radiographer–reporting
Radiographer–reporting
Radiographer–reporting
Radiographer–reporting
Radiologist–reporting
Radiologist–reporting
Radiologist–reporting
Radiologist–reporting
Current intervention
Current intervention
Current intervention
Current intervention
New intervention
New intervention
New intervention
New intervention
Current intervention Radiologist reporting £156 £52.00 per patient/report £34.66 per patient/report £26.00 per patient/report
Noncomplex MRI Report Configuration Cost per Hour CfWI/DoH [12] Slow CfWI/DoH [12] Medium CfWI/DoH [12] Fast
Report (24 min) Report (16 min) Report (12 min)
Current intervention Radiologist reporting £156 £62.40 per patient/report £41.60 per patient/report £31.20 per patient/report
Noncomplex MRI Report Configuration Cost per Hour RCR [36] Slow Report RCR [36] Medium RCR [36] Fast Report
(20 min) Report (13.33 min) (10 min)
New intervention Radiographer reporting £53 £17.66 per patient/report £11.77 per patient/report £8.83 per patient/report
Noncomplex MRI Report Configuration Cost per Hour CfWI/DoH [12] Slow CfWI/DoH [12] Medium CfWI/DoH [12] Fast
Report (24 min) Report (16 min) Report (12 min)
New intervention Radiographer reporting £53 £21.20 per patient/report £14.13 per patient/report £10.60 per patient/report
CfWI, Centre for Workforce Intelligence; DoH, Department of Health; MRI, magnetic resonance imaging; RCR, Royal College of Radiologists.
Social Care Act 2012 [60] for NHS trusts, NHS foundation using the data (n ¼ 3,525) from the acute DGH 12-
trusts, and private providers. month audit of workload. Calculated savings between
Comparison of the interventional cost of reporting ra- £121,048 - £60,524 could be possible using reporting ra-
diographers to report a noncomplex MRI scan against the diographers against the RCR [36] workload model (fast, me-
national tariff of £22 [58] per report demonstrates cost sav- dium, and slow reporting times). Calculating the reporting
ings between £2.83 and £11.17 per scan, calculated against radiographer’s unit costs against the CfWI and DoH [12] re-
all the proposed RCR [36] and CfWI and DoH [12] time porting ranges give an estimated annual cost saving of
ranges to report an MRI scan (Table 7). Extrapolation of £145,230–£72,615 (Table 8) compared with the current
the data allowed approximation over the observed range intervention of radiologists.
Table 8
Potential Unit Costs of per Annum of Current and New Interventions Using DGH Audit of Workload Against the RCR [36], CfWI, and DoH [12] Calculations
Noncomplex MRI Configuration Annual DGH RCR [36] Slow RCR [36] Medium RCR3 [2] Fast Report
Report Workload Report (20 min) Report (13.33 min) (10 min)
Current intervention Radiologist reporting 3,525 MRI scans Annual cost Annual cost Annual cost
£183,300.00 £122,141.25 £91,650.00
Noncomplex MRI Configuration Annual DGH CfWI/DoH [12] Slow CfWI/DoH [12] CfWI/DoH [12] Fast
Report Workload Report (24 min) Medium Report Report (12 min)
(16 min)
Current intervention Radiologist reporting 3,525 MRI scans Annual cost Annual cost Annual cost
£219,960.00 £146,640.00 £109,980.00
Noncomplex MRI Configuration Annual DGH RCR [36] Slow RCR [36] Medium RCR [36] Fast Report
Report Workload Report (20 min) Report (13.33 min) (10 min)
New intervention Radiographer 3,525 MRI scans Annual cost Annual cost Annual cost
reporting £62,251.50 £41,489.25 £31,125.75
Noncomplex MRI Configuration Annual DGH CfWI/DoH [12] Slow CfWI/DoH [12] CfWI/DoH [12] Fast
Report Workload Report (24 min) Medium Report Report (12 min)
(16 min)
New intervention Radiographer 3,525 MRI scans Annual cost Annual cost Annual cost
reporting £74,730.00 £49,808.25 £37,365.00
CfWI, Centre for Workforce Intelligence; DGH, district general hospital; DoH, Department of Health; MRI, magnetic resonance imaging; RCR, Royal Col-
lege of Radiologists.