Cost-effectiveness analysis of the Neuropad device as a screening tool for early diabetic peripheral neuropathy

Abstract

Objective
To carry out a cost-effectiveness analysis of the use of Neuropad as a screening test for diabetic neuropathy together with the standard care tool, the 10-g monofilament, in people with diabetes.

Research design and methods
A cost-effectiveness analysis using a Markov model was developed to assess the impact on costs and outcomes of using Neuropad as a test for diabetic neuropathy (1) as a complement to the standard test, the 10-g monofilament (Neuropad + monofilament vs. monofilament); and (2) as a substitute for the monofilament (Neuropad vs. monofilament); from the healthcare provider perspective. The time horizon was 3 years. Data on costs and health gains were extracted from the literature. The incremental cost–utility ratio was calculated. Deterministic and probabilistic sensitivity analyses were also performed.

Results
Compared with standard care, Neuropad, in combination with the 10-g monofilament tool, is the dominant strategy as it leads to higher health gains and lower costs. In practice, compared with using the monofilament alone, performing both tests would lead to a savings of £1049.26 per patient and 0.044 QALY gain. Results were found to be consistent across the sensitivity analyses.

Conclusions
Using both screening tools (Neuropad + monofilament) is a cost-effective strategy and the dominant alternative, when compared against using the 10-g monofilament alone. The results would be of special relevance in the early detection of diabetic peripheral neuropathy and to ensure the efficient allocation of resources and, thus, the sustainability of healthcare systems.

Full text

1 23
The European Journal of Health
Economics
ISSN 1618-7598
Eur J Health Econ
DOI 10.1007/s10198-019-01134-2
Cost-effectiveness analysis of the Neuropad
device as a screening tool for early diabetic
peripheral neuropathy
B.Rodríguez-Sánchez, L.M.Peña-
Longobardo & A.J.Sinclair

1 23
Your article is protected by copyright and
all rights are held exclusively by Springer-
Verlag GmbH Germany, part of Springer
Nature. This e-offprint is for personal use only
and shall not be self-archived in electronic
repositories. If you wish to self-archive your
article, please use the accepted manuscript
version for posting on your own website. You
may further deposit the accepted manuscript
version in any repository, provided it is only
made publicly available 12 months after
official publication or later and provided
acknowledgement is given to the original
source of publication and a link is inserted
to the published article on Springer's
website. The link must be accompanied by
the following text: "The final publication is
available at link.springer.com”.

Vol.:(0123456789)
1 3
The European Journal of Health Economics
https://doi.org/10.1007/s10198-019-01134-2
ORIGINAL PAPER
Cost‑eectiveness analysis oftheNeuropad device asascreening tool
forearly diabetic peripheral neuropathy
B.Rodríguez‑Sánchez1 · L.M.Peña‑Longobardo1· A.J.Sinclair2
Received: 18 October 2018 / Accepted: 30 October 2019
© Springer-Verlag GmbH Germany, part of Springer Nature 2019
Abstract
Objective To carry out a cost-effectiveness analysis of the use of Neuropad as a screening test for diabetic neuropathy
together with the standard care tool, the 10-g monofilament, in people with diabetes.
Research design and methods A cost-effectiveness analysis using a Markov model was developed to assess the impact on
costs and outcomes of using Neuropad as a test for diabetic neuropathy (1) as a complement to the standard test, the 10-g
monofilament (Neuropad + monofilament vs. monofilament); and (2) as a substitute for the monofilament (Neuropad vs.
monofilament); from the healthcare provider perspective. The time horizon was 3years. Data on costs and health gains were
extracted from the literature. The incremental cost–utility ratio was calculated. Deterministic and probabilistic sensitivity
analyses were also performed.
Results Compared with standard care, Neuropad, in combination with the 10-g monofilament tool, is the dominant strategy
as it leads to higher health gains and lower costs. In practice, compared with using the monofilament alone, performing both
tests would lead to a savings of £1049.26 per patient and 0.044 QALY gain. Results were found to be consistent across the
sensitivity analyses.
Conclusions Using both screening tools (Neuropad + monofilament) is a cost-effective strategy and the dominant alternative,
when compared against using the 10-g monofilament alone. The results would be of special relevance in the early detection
of diabetic peripheral neuropathy and to ensure the efficient allocation of resources and, thus, the sustainability of healthcare
systems.
Keywords Neuropad· Diabetes· Diabetic peripheral neuropathy· Cost-effectiveness· SWME
JEL Classication H00· H5· H51· I00· I1· I10
Introduction
In the United Kingdom (UK), an estimated 4.5 million peo-
ple have diabetes [1], being predicted to rise to 5 million
people by 2025, which would result in higher associated
health complications and premature mortality. Moreover, it
has been estimated that in 2010–2011, the cost of direct
patient care for those living with type 2 diabetes in the UK
was £8.8 billion and the indirect costs were approximately
£13 billion.
Diabetic peripheral neuropathy (DPN) is a common long-
term complication, which consists of nerve damage in hands
and feet, and it also affects internal organs such as heart
and bladder [2, 3]. Alike, DPN affects up to 50% of people
with diabetes [4, 5], with chronic painful neuropathy affect-
ing up to 26%, resulting in higher risk of foot ulceration
and subsequent amputation [6]. In England, around 2.5% of
the population with diabetes, approximately 86,000 people,
have foot ulcers at any given time [7] and there were around
7400 lower limb amputations due to DPN in 2015–2016
[8]. This has led to the fact that in 2014–2015, the National
Electronic supplementary material The online version of this
article (https ://doi.org/10.1007/s1019 8-019-01134 -2) contains
supplementary material, which is available to authorized users.
* B. Rodríguez-Sánchez
beatriz.rsanchez@uclm.es
1 Faculty ofLaw andSocial Sciences, University ofCastilla-
La Mancha, Calle San Pedro Mártir 7, 45002Toledo, Spain
2 Foundation forDiabetes Research inOlder People, Diabetes
Frail Ltd, University ofAston, Birmingham, UK
Author's personal copy

B.Rodríguez-Sánchez et al.
1 3
Health Service (NHS) in England spent an estimated £972
million–£1.13 billion, equivalent to 0.72–0.83% of its entire
budget, on foot ulceration and amputation [9]. In fact, around
two-thirds of this expenditure were on primary care, com-
munity and outpatient settings for ulceration [10].
Due to such issues related to diabetes` complications,
improving screening and earlier identification of patients at
risk of developing the diabetic foot syndrome (DPN) might
offer an excellent opportunity for patients with diabetes. First
of all, because it might be the case that people with diabetes
could not be aware of having DPN since around 50–85% of
cases are asymptomatic [11]. Moreover, the current prac-
tice barely focuses on early detection of DPN, maybe due to
complications related to the diagnosis. Furthermore, if, due
to lack of early diagnosis, people with diabetes and at high
risk of developing neuropathy first present with an ulcer, the
annual cost of treating these patients would be in the region
of £29million–£116million [10]. Furthermore, screening
would also imply behavioural changes to reduce the risk of
unperceived trauma and identify those patients who should
undergo more intense intervention including improved gly-
caemic, blood pressure and lipid control and, if at particu-
larly high risk, referral to multidisciplinary foot care teams
[12, 13].
However, current primary care tests for neuropathy (such
as Neuropathy Disability Score (NDS), Tuning forks and
Semmes–Weinstein monofilament examination 10-g) are
subjective, since they require the patients’ response, these
tools have a high probability of reporting both false negative
and false positive results. Despite the frequent use of the
Semmes–Weinstein monofilament examination (SWME),
little can be said about the test accuracy for detecting neu-
ropathy in feet without visible ulcers [14, 15]. According
to the National Institute for Health and Care Excellence
(NICE), false positives are preferable to false negatives lead-
ing to the recommendation of 10-g monofilament, despite
the lack of evidence supporting its use [16]. Some authors
motivate the use of complementary diagnostic or screening
tests together with the monofilament testing [14], such as
Neuropad, which is a simple sudomotor function test (SFT)
[17]. Neuropad is the only self-testing device for sudomotor
function available for use in a primary care or the home set-
ting. More specialist tests, for example NDS, are used in sec-
ondary care to detect small fibre neuropathy. Therefore, for
a firm diagnosis of diabetic autonomic neuropathy (DAN),
patients identified as at potential risk should be referred to
secondary care where further hospital-based tests may be
carried out to confirm the diagnosis, as Fig.1 shows.
Early detection might also play a relevant role with
respect to the health-related quality of life (HRQoL) on
people with diabetes additionally suffering of DPN. In fact,
the negative association between DPN and HRQoL has
been supported by several studies [18, 19], highlighting the
relevance of such relationship in case of painful diabetic
peripheral neuropathy. Moreover, it has been supported that
quality of life in people with current ulcers and for those
People with diabetes attending annual
diabetes care review
Routine assessment for preclinical DPN
at diabetes annual review using
Neuropad screening test
LOW RISK
Also consider using a 10g
monofilament to rule out
evidence of clinical DPN
Abnormal –blue or mixed
blue/pink result: anhidrosis
detected
Normal –pink result:
anhidrosis not detected
Screen for sensory neuropathy
using a 10g monofilament
Normal result:
Protective sensation
retained
Abnormal result:
protective sensation
poor or absent
HIGH RISK
Clinical DPN detected.
Protective foot
sensation lost. Notify
footcare
MDT/specialist team.
MODERATE RISK
Pre-clinical DPN detected.
Inform diabetes healthcare
teamso they can advise on
patient’s metabolic control.
Fig. 1 Diabetes annual foot examination flowchart. Source: Author´s elaboration
Author's personal copy

Cost‑effectiveness analysis oftheNeuropad device asascreening tool forearly diabetic…
1 3
with major amputations is lower than for individuals with
other long-term diseases such as chronic obstructive pulmo-
nary disease or renal disease [10].
Due to all the previous issues mentioned before, it could
be quite important an earlier diagnosis. In this sense, Neu-
ropad could allow patients with neuropathy to be diagnosed
earlier than is possible with current tests, allowing clinicians
to target and triage those patients who should undergo more
intense multifactorial intervention. Neuropad is a non-sub-
jective test with a sensitivity, specificity and reliability com-
parable to established secondary care diagnostics [20–24].
Since the accuracy of foot risk assessment tools to predict
ulceration also requires the need for economic assessments
to value not only how effective a diagnostic tool is, but also
how costly it can get compared to the standard of care [25],
this paper provides with the cost-effectiveness analysis of
Neuropad for patients with diabetes and at risk of suffering
from peripheral neuropathy from the healthcare provider
perspective of the UK NHS. Neuropad will be compared
with current standard care, the SWME, both individually
and jointly: Neuropad versus SWME; and Neuropad together
with SWME versus SWME.
Research design andmethods
The cost-effectiveness of Neuropad from a healthcare pro-
vider perspective was determined using a Markov model1
(Fig.1) to simulate the health and economic outcomes of
foot care in a hypothetical population of people with dia-
betes. Markov models are particularly useful in economic
evaluations of progressive chronic conditions [27], as it
is the case of diabetes and diabetic foot disease. Within
Markov models, individuals are allocated into health states,
with each state having specific costs and health outcomes,
operating in cycles. People remain in each health state for
one cycle and can progress to a separate state at the end of
the cycle or remain in the same state. The population mod-
elled in this economic evaluation would be any person of any
age with diabetes (type 1 diabetes, type 2 diabetes, and rarer
types) without a prior diagnosis of peripheral neuropathy, an
active ulcer, a previous ulcer, a previous amputation, or other
causes such as low levels of vitamin B12, kidney disease,
and thyroid problems. This would be around 80% of patients,
as 20% will have a prior diagnosis of neuropathy.
The current model contains seven health states: no neu-
ropathy, neuropathy, infected foot ulcer, minor amputation,
major amputation, healed foot and death. Expert clinical
advisers were consulted for their approval on the disease
progression model.2 Cycle length in this analysis is of 6
months and the model covers a 3-year period, following
some literature found in the cost-effectiveness analysis of
diabetic neuropathy diagnostic or control tools [28, 29].
Model structure
State A represents the healthiest individuals, with no signs
of diabetic neuropathy (so being healthy does not mean no
disease, but no neuropathy), but still suffering from diabetes.
State B refers to those who have been diagnosed of neuropa-
thy (both true positives and false positives considered), and
State C includes patients who already have an infected foot
ulcer. States D and E refer to patients who have progressed
to minor and major, respectively, lower limb amputation.
State F refers to ulcers that have been healed. Although it
is not shown in Fig.2, the analysis also takes into account
the chance of death for all health states. The arrows in the
model show how patients can progress through the model
over the cycles.
Patients progressed to other health states, remained in
the same state, or died, depending on the associated prob-
abilities for each transition. Six-month transition probabili-
ties were assigned for movement between the health states.
Individuals entering the model can undergo different tests:
Neuropad or SWME. If Neuropad is used and an abnormal
result is obtained, they would undergo the SWME test as
well or not. Hence, three strategies will be evaluated in the
analysis: Neuropad alone, SWME alone, and Neuropad with
SWME.
Most of the parameters were sourced from Ortegon etal.
[29] and Ragnarson-Tennvall and Apelqvist [30], although
a more detailed description of all the values applied in the
model together with the source where they have been found
is provided in Table1.
To keep the model tractable, a number of assumptions are
made. All patients are tested prior to entry into the model
and placed in one of the following four health outcomes after
each screening/diagnostic test, which are expected to take
into account potential mistakes due to the tests´ specificity
1 The Markov model was constructed using Microsoft Excel (Micro-
soft Corporation, Redmond, WA, USA) to emulate the different clini-
cal pathways that can be developed once a person with diabetes has
been diagnosed with neuropathy [26].
2 We did not define comorbidities (as we neither defined age, nation-
ality and functional status) as the inclusion of these would have com-
plicated the analysis unnecessarily and made the interpretation of the
findings very difficult and unlikely to allow us to find clear conclu-
sions. Any of the seven health states considered could apply to people
with diabetes of any age since as many as 20% of newly diagnosed
people with type 2 diabetes have signs of neuropathy at diagnosis.
However, those with an ulcer or a history of amputation are likely to
have had diabetes for at least 10years, irrespective of the duration of
neuropathy remembering that vascular disease may be the more pre-
dominant cause.
Author's personal copy

B.Rodríguez-Sánchez et al.
1 3
and sensitivity levels: No DPN (true negative), DPN (true
positive), DPN (false negative), and false positive. Further
assumptions include: population mortality is independent of
age; patient survival is considered to be the same regardless
of the test choice; all patients enter the foot care programme
after ulceration occurs; patients testing positive for DPN
join the foot care programme and no further tests for DPN
are undertaken; and once a major amputation has occurred,
ulceration of the ipsilateral foot does not occur.
In the simulations, testing for peripheral diabetic neu-
ropathy with Neuropad alone and with Neuropad and 10-g
SWME was compared with the present level of care (10-g
SWME only).
Costs
The economic outcomes studied were incremental cost per
person with diabetes and the Incremental Cost–Utility Ratio
(ICUR). All costs are expressed in 2015pounds.
As it is shown in Table1, the purchase price of both
Neuropad and SWME will be considered in the analysis.
Some assumptions regarding the diagnostic tests (Neuro-
pad and 10-g SWME) costs selection should be mentioned:
associated costs (staff time/training/infrastructure) have not
been included in the cost analysis since no costs associated
with Neuropad have already been reported in the literature.
SWME has indeed been supported to require trained health-
care professional to perform the test [31], but those esti-
mates do not provide with certain information about how
much staff time need to be used to interpret an abnormal
result of the test. Cost per patient/use was also neglected
for both screening tools. It is uncertain how many times 1
monofilament can be used: some report they are reusable;
some other state they should be replaced between patients
to reduce infections [31].
Resource use associated with the various health states
and transitions was obtained from a published study on dia-
betic foot care costs in England [10], from another cost-
effectiveness analysis on diabetic foot [28] and from an
economic study on the treatment of diabetic foot ulcers and
amputations [30]. Only direct medical costs were considered
(Table1). It should be noted that cost of stumps of ampu-
tations will be included not only in the cycle at which the
amputation takes place, but also in the succeeding cycles as
a cost of care for amputations.
A discount rate of 3.5% is applied to all costs beyond the
first year [32].
Utilities
The clinical outcome measured was the incremental quality-
adjusted life years (QALYs). QALYs were used as the health
outcome for being considered as a life and health measure
[33]. Outcomes from four different studies [28–30, 34] were
used for entering the utilities into the cost-effectiveness
model (Table1). A discount rate of 3.5% is applied to all
QALYs beyond 1year as costs since health gains are also
tradable [32, 35].
Analysis
The decision rule for the optimal strategy was based on the
incremental cost–utility ratio (ICUR). It should be noted
that when comparing the ICUR to the threshold, estima-
tions of the same sign could denote completely different
results. For example, a positive ICUR might represent
the assessed intervention, against its comparator, is more
Fig. 2 Markov model structure
on Health states and transi-
tion paths. Source: Author´s
elaboration State A
No neuropathy
State B
Neuropathy
State C
Infected foot ulcer
State D
Minor amputation
State E
Major amputation
State F
Healed
State G
Death
Author's personal copy

Cost‑effectiveness analysis oftheNeuropad device asascreening tool forearly diabetic…
1 3
Table 1 List of parameters used in the analysis and their source
Variable Value 95% CI range Source
Utility values assigned to specific health states *. 95% CI given, if available
No active ulcer–no previous amputation 0.84 (0.81, 0.87) Redekop etal. (2004)
No active ulcer–only 1 + toes amputated 0.74 (0.70, 0.78) Redekop etal. (2004)
No active ulcer–one foot amputated 0.68 (0.63, 0.72) Redekop etal. (2004)
No active ulcer–one leg amputated 0.62 (0.57, 0.67) Redekop etal. (2004)
No active ulcer–both feet or legs amputated 0.51 (0.46, 0.55) Redekop etal. (2004)
Active uninfected ulcer–no previous amputation 0.75 (0.71, 0.79) Redekop etal. (2004)
Active uninfected ulcer–only 1 + toes amputated 0.68 (0.64, 0.73) Redekop etal. (2004)
Active uninfected ulcer–one foot amputated 0.63 (0.59, 0.68) Redekop etal. (2004)
Active uninfected ulcer–one leg amputated 0.57 (0.53, 0.62) Redekop etal. (2004)
Active infected ulcer–no previous amputation 0.70 (0.66, 0.75) Redekop etal. (2004)
Active infected ulcer–only 1 + toes amputated 0.65 (0.60, 0.69) Redekop etal., 2004
Active infected ulcer–one foot amputated 0.59 (0.54, 0.63) Redekop etal. (2004)
Active infected ulcer–one leg amputated 0.55 (0.50, 0.59) Redekop etal. (2004)
No neuropathy 0.84 NR Ortegon etal. (2004)
Neuropathy 0.74 NR Ortegon etal. (2004)
After healing with minor amputation 0.61 NR Ragnarson-Tennvall and Apelqvist (2001)
Transition probabilities between health states
Sensitivity Neuropad (%) 86 (79, 91) Tsapas etal. (2014)1
Specificity Neuropad (%) 65 (51, 76) Tsapas etal. (2014)1
Sensitivity SWME (%) 84 NR Willits etal. (2015)2
Specificity SWME (%) 83 NR Willits etal. (2015)2
Sensitivity Neuropad + SWME (%) 75 NR Calculated: SensitivityNeuropad * SensitivitySWME
Specificity Neuropad + SWME (%) 93 NR Calculated: SpecificityNeuropad + (1 − SpecificityNeuropad *
SpecificitySWME)
Neuropathy prevalence (%) 2.4 NR Kostev etal. (2014)3
Minor amputations prevalence (%) 57.10 NR Kerr (2017)
Major amputations prevalence (%) 42.90 NR Kerr (2017)
No neuropathy–no neuropathy (%) 95.34 NR Estimated
No neuropathy–neuropathy (%) 1.99 NR Ortegon etal. (2004)
No neuropathy–infected foot ulcer (%) 0.26 NR Ortegon etal. (2004)
Neuropathy–neuropathy (%) 96.60 NR Estimated
Neuropathy–infected foot ulcer (%) 1.40 NR Ragnarson-Tennvall and Apelqvist (2001)
Infected foot ulcer–infected foot ulcer (%) 36.00 NR Estimated
Infected foot ulcer–minor amputation (%) 13.00 NR Prompers etal. (2008)4
Infected foot ulcer–major amputation (%) 5.00 NR Prompers etal. (2008)4
Infected foot ulcer–healing (%) 40.00 NR Ragnarson-Tennvall and Apelqvist (2001)
Minor amputation–neuropathy (%) 9.60 NR Ortegon etal. (2004)
Transition probabilities between health states
Minor amputation–infected foot ulcer (%) 7.3 NR Ragnarson-Tennvall and Apelqvist (2001 ) (including
ulcers with critical ischaemia)
Minor amputation–minor amputation (%) 63.40 NR Estimated
Minor amputation–major amputation (%) 17.00 NR Ortegon etal. (2004)
Major amputation–major amputation (%) 88.00 NR Estimated
Healing–infected foot ulcer (%) 7.30 NR Ragnarson-Tennvall and Apelqvist (2001) (including
ulcers with critical ischaemia)
Healing–healing (%) 90.00 NR Estimated
No neuropathy–Death (%) 2.00 NR Ortegon etal. (2004)
Neuropathy–Death (%) 2.00 NR Assumed same as No neuropathy
Infected foot ulcer–Death (%) 6.00 NR Prompers etal. (2008)4
Author's personal copy

B.Rodríguez-Sánchez et al.
1 3
efficient and with higher costs or less efficient but leads
to lower costs. Moreover, a negative ICUR could mean
that the new alternative is less costly and more effective
in terms of health gains and, hence, the new alternative
would be dominant and preferred over usual care. But a
negative ICUR could also reflect that the new intervention
would never be adopted as it results in higher costs and
lower health gains. If a negative ICUR is reported denoting
health gains and cost savings, the strategy would dominate.
Likewise, if a positive ICUR below the threshold value is
shown, it would also be cost-effective and dominant.
NR not reported
*QALYs and costs discounted in cycles 3, 4, 5 and 6
a Hospital Episode Statistics for England, national tariffs and NHS Reference Costs were used to estimate inpatient activity and costs. NHS Ref-
erence costs were also used to estimate outpatient costs. Staff unit costs were taken from the Personal Social Services Research Unit, and the cost
of medications from the NHS Electronic Drug Tariff and the British National Formulary
b Costs are expressed as 2015 prices
**Costs were originally Euros, so they have been converted to pounds and 2015 prices
1 Tsapas, A.; Liakos, A.; Paschos, P.; Karagiannis, T.; Bekiari, E.; Tentolouris, N.; Boura, P. (2014) A simple plaster for screening for diabetic
neuropathy: a diagnostic test accuracy systematic review and meta-analysis. Metabolism Clinical and Experimental; 63:584–592
2 Willits, I., etal. (2015) VibraTipTM for testing vibration perception to detect diabetic peripheral neuropathy: a NICE medical technology guid-
ance. Applied Health Economics and Health Policy; 13(4):315–324
3 Kostev, K.; Jockwig, A.; Hallwachs, A.; Rathmann, W. (2014) Prevalence and risk factors of neuropathy in newly diagnosed type 2 diabetes in
primary care practices: a retrospective database analysis in Germany and UK. Primary Care Diabetes; 8(3):250–255
4 Prompers, L.; Schaper, N.; Apelqvist, J.; Edmonds, M.; Jude, E.; Mauricio, D.; Uccioli, L.; Urbancic, V.; Bakker, K.; Holstein, P.; Jirkovska,
A.; Piaggesi, A.; Ragnarson-Tennvall, G.; Reike, H.; Spraul, M.; Van Acker, K.; Van Baal, J.; Van Merode, F.; Ferreira, I.; Huijberts, M. (2008)
Prediction of outcome in individuals with diabetic foot ulcers: focus on the differences between individuals with and without peripheral arterial
disease. The EURODIALE Study. Diabetologia; 51(5):747–55
5 Claxton, K.; Sculpher, M.; McCabe, C.; Briggs, A.; Akehurst, R.; Buxton, M.; Brazier, J.; O’Hagan, T. (2005) Probabilistic sensitivity analysis
for NICE technology assessment: not an optional extra. Health Economics; 14(4):339–47
Table 1 (continued)
Variable Value 95% CI range Source
Minor amputation–Death (%) 2.7 NR Ragnarson-Tennvall and Apelqvist (2001)
Major amputation–Death (%) 12.00 NR Ragnarson-Tennvall and Apelqvist (2001)
Healing–Death (%) 2.7 NR Ortegon etal. (2004)
Costs applied in the economic analysis*,a,b
6months cost per patient of primary and community
care if neuropathy
£1855.92 NR Kerr (2017)
6months cost per patient of primary and community
care if infected foot ulcer
£8620.8 NR Kerr (2017)
6months cost per patient of inpatient care for minor
amputations
£2105.89 NR Kerr (2017)
6months cost per patient of inpatient care for major
amputations
£4106.85 NR Kerr (2017)
Costs applied in the economic analysis*,a,b
6months cost per patient of inpatient care for proce-
dures on stumps
£2812.30 NR Kerr (2017)
6months cost per patient of inpatient care for foot
ulcers
£3227.27 NR Kerr (2017)
6months cost per patient of no neuropathy £125,04 NR Green and Taylor (2016)
6months cost per patient of healing £125,04 NR Assumed to be equal to the cost of no neuropathy
Transition cost from infected foot ulcer to amputa-
tion**
£9407 (£3395–£74,387) Ragnarson-Tennvall and Apelqvist (2001)
Purchase price of Neuropad £7.28 NR NICE (2016)
Purchase price of 10-g SWME £16.80 NR NICE (2016)
Discount rate
Discount rate (%) for any costs or QALYs beyond 1
year
3.5 Claxton etal. (2006, 2011)5
Author's personal copy

Cost‑effectiveness analysis oftheNeuropad device asascreening tool forearly diabetic…
1 3
The incremental cost–utility ratio responds to the follow-
ing expression [36]:
where
Ci
is the cost of the alternative i compared to the cost
of alternative b,
Cb
, which would be current practice (10-g
SWME).
QALYSi
denotes the health benefits of alternative i
compared to the health gains of alternative b,
QALYSb
. The
resulting ICUR will be compared against the threshold value
used in this analysis. The threshold can be defined as the
numeric value that society is willing to pay to increase the
health benefit in one unit. Some uncertainty surrounded the
cost per QALY proposed by NICE, which was supposed to
be within the range £20,000–£30,000 [37], using the thresh-
old lower bound of £20,000 in the current analysis.
Sensitivity analysis
A deterministic sensitivity analysis (DSA), both one-way
and two-way sensitivity analyses, was performed among
the following variables: costs associated with the different
health states; purchase price of Neuropad and SWME; sen-
sitivity and specificity of Neuropad and SWME; discount
rate; peripheral neuropathy prevalence; QALYs associated
with the different health states; and transition probabilities
between health states. The major goal was testing which
parameters were those with the greatest impact on the ICUR,
whether the optimal strategy changes when the parameters
are modified and, consequently, whether the lower or upper
bound of such influencing variables lead to relevant changes
in the ICUR.
All parameters entered in the simulation model were
assumed to increase or decrease by 20% (QALYs and costs,
transition probabilities, sensitivity and specificity of both
Neuropad and SWME, and prevalence of neuropathy) [27,
36]. Hence, the effects of an over- or underestimation in any
of the parameters included in the study would also be tested.
As care home residents are at greater risk of develop-
ing diabetic neuropathy and no appropriate care and foot
screening is provided [38], it seems reasonable to take
such group of the population into consideration when
evaluating Neuropad as a diagnostic tool of diabetic neu-
ropathy. However, it should be mentioned that, due to lack
of data about the diabetic neuropathy prevalence within
UK care home residents, no actual prevalence of diabetic
neuropathy could be used in the analysis. Nevertheless, it
has been found in the literature that for a subsample of 497
Dutch care home residents, the prevalence of actual neu-
ropathy pain was 10.9% (95% CI 8.4–13.8%) and 7.7% for
the residents suffering from diabetes [39]. Another study
reported higher estimates on the prevalence of painful
ICUR
i=
(
Ci−Cb
)
(
QALYSi−QALYSb
),
diabetic peripheral neuropathy in the UK found it to be
26.4% [40]. Hence, diabetic neuropathy prevalence will
be modified in the current analysis from its value for the
overall sample (2.4%) up to 27%%, which is above the
upper bound of the 95% CI provided in the Dutch paper
on care home residents.
Although deterministic sensitivity analysis remains
the most popular technique to account for uncertainty, it
gives insufficient insight into the scale of decision uncer-
tainty [36]. Therefore, a probabilistic sensitivity analysis
(PSA) was performed as a complementary approach that
involves specifying distributions for input parameters and
employing Monte Carlo simulations, allowing the joint
effect of parameter uncertainty to be assessed [37]. The
probabilistic analysis was undertaken by randomly sam-
pling each of the parameter distributions and calculating
the expected costs and outcomes for that combination of
parameter values. This process formed a single replication
of the model results, and a total of 10,000 iterations were
performed to examine the distribution of the resulting cost
and outcomes for each intervention. For costs calculations,
the distribution used for calculating the random interac-
tions was a Gamma distribution, while a beta distribution
was used for QALYs and transition probabilities [36, 37].
Cost-effectiveness acceptability curves were also con-
structed to assess decision uncertainty, where the prob-
ability of being the most cost-effective strategy was plotted
against cost-effectiveness threshold values ranging from
£0 to £100,000 for the strategy Neuropad and the 10-g
SWME together versus using only the 10-g SWME..
Table2 shows how the model parameters were modi-
fied in the deterministic and the probabilistic sensitivity
analyses.
Results
Health andEconomic outcomes
Table3 shows that, compared with the standard of care
(SWME 10-g alone), Neuropad together with SWME is the
dominant strategy (negative ICUR), as it leads to higher
health gains and lower costs (southeast quadrant of the
cost-effectiveness plane). Compared to using SWME only
as the diagnostic test, performing both tests would lead to
a £1049.26 savings per patient and 0.044 QALY gain. On
the other hand, the individual use of Neuropad, compared
to the monofilament, seems to be a dominant strategy, as
it leads to lower health gains and higher costs (northwest
quadrant). In fact, using Neuropad alone, against its com-
parator, derives into £1936.01 more and 0.070 QALYs
less.
Author's personal copy

B.Rodríguez-Sánchez et al.
1 3
Table 2 Changes in model parameters for the deterministic and probabilistic sensitivity analysis
Variable Basecase value Deterministic sensitivity analysis Probabilistic
sensitivity
analysis
− 20% [If a variation of − 20% led to
a non-sense results (i.e., a value below
zero), we took 0 as the maximum lower
bound.]
+ 20% [If a variation of + 20% derived
into a probability or utility value above 1,
we used 1 as the maximum upper bound
of change.]
Distribution
Utilities
No active ulcer–no previous amputation 0.84 67.20 100.00 Beta
No active ulcer–only 1 + toes amputated 0.74 59.20 88.80 Beta
No active ulcer–one foot amputated 0.68 54.40 81.60 Beta
No active ulcer–one leg amputated 0.62 49.60 74.40 Beta
No active ulcer–both feet or legs ampu-
tated
0.51 40.80 61.20 Beta
Active uninfected ulcer–no previous
amputation
0.75 60.00 90.00 Beta
Active uninfected ulcer–only 1 + toes
amputated
0.68 54.40 81.60 Beta
Active uninfected ulcer–one foot ampu-
tated
0.63 50.40 75.60 Beta
Active uninfected ulcer–one leg ampu-
tated
0.57 45.60 68.40 Beta
Active infected ulcer–no previous amputa-
tion
0.70 56.00 84.00 Beta
Active infected ulcer–only 1 + toes
amputated
0.65 52.00 78.00 Beta
Active infected ulcer–one foot amputated 0.59 47.20 70.80 Beta
Active infected ulcer–one leg amputated 0.55 44.00 66.00 Beta
No neuropathy 0.84 67.20 100.00 Beta
Neuropathy 0.74 59.20 88.80 Beta
After healing with minor amputation 0.61 48.80 73.20 Beta
Sensitivity and specificity of screening/diagnostic tools and Transition probabilities between health states (%)
Sensitivity Neuropad 86 68.80 100.00 Beta
Specificity Neuropad 65 52.00 78.00 Beta
Sensitivity SWME 84 67.20 100.00 Beta
Specificity SWME 83 66.40 99.60 Beta
Sensitivity Neuropad + SWME 75 60.00 90.00 Beta
Specificity Neuropad + SWME 93 74.40 100.00 Beta
Neuropathy prevalence 2.4 1.92 2.86 Beta
Author's personal copy

Cost‑effectiveness analysis oftheNeuropad device asascreening tool forearly diabetic…
1 3
Table 2 (continued)
Variable Basecase value Deterministic sensitivity analysis Probabilistic
sensitivity
analysis
− 20% [If a variation of − 20% led to
a non-sense results (i.e., a value below
zero), we took 0 as the maximum lower
bound.]
+ 20% [If a variation of + 20% derived
into a probability or utility value above 1,
we used 1 as the maximum upper bound
of change.]
Distribution
No neuropathy–no neuropathy 95.34 76.27 100.00 Beta
No neuropathy–neuropathy 1.99 1.592 2.388 Beta
No neuropathy–infected foot ulcer 0.26 0.208 0.312 Beta
Neuropathy–neuropathy 96.60 77.28 100.00 Beta
Neuropathy–infected foot ulcer 1.40 1.12 1.68 Beta
Infected foot ulcer–infected foot ulcer 36.00 28.80 43.20 Beta
Infected foot ulcer–minor amputation 13.00 10.40 15.60 Beta
Infected foot ulcer–major amputation 5.00 4.00 6.00 Beta
Infected foot ulcer–healing 40.00 32.00 48.00 Beta
Minor amputation–neuropathy 9.60 7.68 11.52 Beta
Minor amputation–infected foot ulcer 7.3 5.84 8.76 Beta
Minor amputation–minor amputation 63.40 50.72 76.08 Beta
Minor amputation–major amputation 17.00 13.60 20.40 Beta
Major amputation–major amputation 88.00 70.40 100.00 Beta
Healing–infected foot ulcer 7.30 5.84 8.76 Beta
Healing–healing 90.00 72.00 100.00 Beta
No neuropathy–Death 2.00 1.60 2.40 Beta
Neuropathy–Death 2.00 1.60 2.40 Beta
Infected foot ulcer–Death 6.00 4.80 7.20 Beta
Minor amputation–Death 2.70 2.16 3.24 Beta
Major amputation–Death 12.00 9.60 14.40 Beta
Healing–Death 2.70 2.16 3.24 Beta
Costs (in 2015£)
Purchase price of Neuropad £7.28 £5.824 £8.736 Gamma
Purchase price of 10-g SWME £16.80 £13.44 £20.16 Gamma
6months cost per patient of primary and
community care if neuropathy
£1855.92 £1484.736 £2227.104 Gamma
Author's personal copy

B.Rodríguez-Sánchez et al.
1 3
Table 2 (continued)
Variable Basecase value Deterministic sensitivity analysis Probabilistic
sensitivity
analysis
− 20% [If a variation of − 20% led to
a non-sense results (i.e., a value below
zero), we took 0 as the maximum lower
bound.]
+ 20% [If a variation of + 20% derived
into a probability or utility value above 1,
we used 1 as the maximum upper bound
of change.]
Distribution
6months cost per patient of primary,
inpatient and community care if infected
foot ulcer
£11,840.07 [£11,840.07 is the sum of the
6months cost per patient of primary
and community care if the person with
diabetes has an infected foot ulcer
(£8620.8) plus the 6months cost per
patient of inpatient care for foot ulcers
(£3227.27), as reported in Table1.]
£9478.432 £14,217.648 Gamma
6months cost per patient of inpatient care
for minor amputations
£2105.89 £1684.712 £2527.068 Gamma
6months cost per patient of inpatient care
for major amputations
£4106.85 £3285.48 £4928.22 Gamma
6months cost per patient of inpatient
care for procedures on stumps of minor
amputations [The cost of stumps for
minor amputations was estimated as the
6months cost per patient of inpatient
care for procedures on stumps * minor
amputations prevalence, according to
the values shown in Table1.]
£1605.9429 £1284.654 £1927.132 Gamma
6months cost per patient of inpatient
care for procedures on stumps of major
amputations [The cost of stumps for
major amputations was estimated as the
6months cost per patient of inpatient
care for procedures on stumps * major
amputations prevalence, according to
the values displayed in Table1.]
£1206.3571 £965.086 £1447.629 Gamma
6months cost per patient of no neuropa-
thy
£125.04 £100.032 £150.048 Gamma
6months cost per patient of healing £125.04 £100.032 £150.048 Gamma
Transition cost from infected foot ulcer to
amputation
£9407 £7525.6 £11,288.4 Gamma
Discount rate
Discount rate (%) for any costs or QALYs
beyond 1year
3.5 2.8 4.2 Beta
Author's personal copy

Cost‑effectiveness analysis oftheNeuropad device asascreening tool forearly diabetic…
1 3
Sensitivity analysis
Deterministic sensitivity analysis (DSA)
Both the one-way and two-way DSA show that the
results are quite robust, irrespective of the change of the
parameters.
Moreover, as Fig.3 shows, utility of neuropathy, the
transition probability from neuropathy to neuropathy state,
and the utility associated with no neuropathy are the most
sensitive variables modifying the ICUR.
One‑way sensitivity analysis The conclusions derived from
the one-way sensitivity analysis are that performing Neuro-
pad together with 10-g SWME is always the optimal choice,
compared to using the 10-g SWME alone.
The three most sensitive ICUR drivers are:
• Utility of neuropathy: when the value of the utility
derived from not having neuropathy, but still being dia-
betic, varies by 20%, so between 0.592 and 0.888, the
ICUR, although always negative pointing towards Neu-
ropad + SWME being the dominant alternative, changes
to £− 13,319.94 and £− 1323,900.4, respectively.
Table 3 Expected and incremental health gains (QALYs), costs and incremental cost–utility ratio (ICUR) of the three testing strategies. Source:
Author’s elaboration
E(QALYs) denote the expected QALYs obtained for each strategy from the simulation model; E(costs) represents the expected costs for each
strategy; and I(QALYs) and I(Costs) are the incremental QALYs and costs, respectively, of Neuropad + SWME or Neuropad alone, compared to
the standard of care, which is using SWME only
Strategy E(QALYs) E(Costs) I(QALYs) I(Costs) ICUR
Neuropad + SWME 2.40 £1537.63 0.04 £− 1049.26 Dominant
Neuropad 2.29 £4522.90 − 0.070 £1936.01 Dominated
SWME 2.36 £2586.89 – – –
Fig. 3 Tornado diagram for the ten parameters with the largest impact on the variation of incremental cost–utility ratio. Source: Author´s elabo-
ration
Author's personal copy

B.Rodríguez-Sánchez et al.
1 3
• The transition probability of being in the neuropathy
health state in one cycle and remain in the same state in
the following cycle period: if the probability varies by
20%, so between 0.7728 and 1, the ICUR, which always
points to Neuropad and the monofilament as the domi-
nant strategy, changes to £− 2167.94 and £− 331,834.75,
respectively.
• Utility of no neuropathy: when the value of the utility
derived from not having neuropathy, but still being dia-
betic, varies by 20%, so between 0.672 and 1, the ICUR
changes to £116,431.14 and £− 12,164.71, respectively.
The joint use of Neuropad and the 10-g monofilament
is not cost-effective until the utility of staying in the no
neuropathy state is, at least, equal to 70.48. From this
value up to 1, the use of both tests is always the dominant
strategy.
Two‑way sensitivity analysis Similar findings are obtained
from the two-way sensitivity analysis, compared to the
one-way sensitivity analysis. The aim was to test whether
changes in more than one parameter at the same time would
lead to a change in the optimal strategy (testing diabetic
neuropathy with Neuropad and the 10-g SWME). Looking
at the figuresA1–A10 in Appendix, performing the Neuro-
pad test together with the 10-g SWME is always the optimal
choice, compared to performing 10-g SWME alone.
Subgroup analysis When the prevalence of diabetic neu-
ropathy is modified from its original value (2.4%) to a maxi-
mum of 27%, costs, QALYs and the incremental cost–utility
ratio change (Table4).
• When the prevalence is modified, testing diabetic neu-
ropathy with Neuropad and 10-g SWME stands always
as the optimal strategy.
• Costs increase from £1538.29 as it is in the baseline value
to £3246.50 in case of a prevalence of 27%.
• QALYs decrease from 2.40 to 2.34.
Probabilistic sensitivity analysis (PSA)
Figure4 shows the net incremental cost and net incremen-
tal effectiveness of using Neuropad + SWME compared
to standard care (SWME) after Monte Carlo simulations.
Figure4 shows an incremental average costs of using Neu-
ropad + SWME compared to the use of only SWME of
£1.049.26 and an average of differences on QALYs of 0.044,
suggesting that the use of Neuropad and the 10-g SWME is
a cost-effective strategy for the threshold value of £20,000
and, thus, preferred over the standard care. These results
are confirmed by the cost-effectiveness acceptability curve,
which was not included in the manuscript because there was
a 100% of probability of Neuropad + the 10-g SWME being
dominant for whatever willingness to pay threshold. On the
Table 4 Changes in QALYs, costs, and ICUR when diabetic peripheral neuropathy (DPN) is modified in the subgroup analysis. Source:
Author´s elaboration
DPN prevalence 0.024 0.092 0.1347 0.1655 0.2208 0.27
QALYs 2.40 2.39 2.37 2.37 2.35 2.34
Costs 1538.29 2008.05 2306.99 2520.51 2904.86 3246.50
ICUR − 26,374,53 − 26,367.18 − 26,362.53 − 26,359.23 − 26,353.32 − 26,348.09
Fig. 4 Results from the
probabilistic sensitivity analysis
(PSA). Neuropad + SWME ver-
sus SWME. Source: Author´s
elaboration
Author's personal copy

Cost‑effectiveness analysis oftheNeuropad device asascreening tool forearly diabetic…
1 3
other hand, Neuropad alone was never cost-effective com-
pared with the 10-g SWME.
Discussion
Our results show that using Neuropad (as a sudomotor func-
tion test, SFT) together with SWME as screening and diag-
nostic tests of diabetic peripheral neuropathy is an optimal
strategy (dominant alternative), compared with the current
approved approach of using a 10-g SWME alone, leading
to cost savings and health gains. In fact, using both DPN
screening and diagnostic tools lead to £1049.26 savings per
patient and 0.044 QALY gain. Hence, the first conclusion
that can be drawn is that testing different components of
neuropathy with different technologies (sudomotor func-
tion with Neuropad and feeling/sensation in the feet with
10-g SWME) is not going to increase costs; actually, costs
would be reduced. To our knowledge, this study is the first
economic analysis comparing the use of Neuropad with the
accepted standard approach (10-g SWME) for the detection
of peripheral neuropathy in patients with diabetes. In the
absence of any previous studies using a test of sudomotor
function in this way, it is not possible to make any compara-
tive conclusions.
Our findings suggest that it would be logical for most of
the European health systems to utilise an SFT in conjunc-
tion with the standard care, given the increased prevalence
of diabetic peripheral neuropathy [41] and the limitations in
the early detection of such disease. This could be done by
mailing a sample test to people with diabetes or asking them
to pick the test up from a community pharmacy, for example.
Those that test positive could then be additionally tested
using SWME, if convenient, or even referred to secondary
care for a firm diagnosis.
The conclusion derived from the analysis seems to be
robust, as it has already been shown in the sensitivity analy-
sis. However, it has also been noted that some parameters
seem to play a key role in the cost-effectiveness analysis.
These are mainly the utility associated with neuropathy and
no neuropathy, as well as the transition probability from
neuropathy to neuropathy state. The former two parameters
point out the relevance in terms of quality of life of avoiding
DPN and the need for a proper management of neuropathy
to develop more severe complications.
Nevertheless, a number of limitations should also be men-
tioned. Other complications that were derived from diabetes
and could increase the care costs were not included in the
model. As it has been mentioned in the “Research design
and Methods section”, we did not include other comor-
bidities apart from the ones considered in the seven health
states as their inclusion would have complicated the analysis
unnecessarily and made the interpretation of the findings
very difficult and unlikely to allow us to find clear conclu-
sions. However, we do accept this issue as a limitation, since
if the inclusion of these various additional descriptive factors
had been possible, the findings might have added to the clini-
cal relevance and validity of our methodological approach.
However, costs were varied in the sensitivity analysis by
20% and the results were quite robust to changes (either
decrease or increase) in costs. One of the assumptions of
the model was that SWME cost only referred to the pur-
chase price but did not include consumable costs or required
trained staff, which is needed but it is not known ascertained
how much. However, SWME cost was also modified in the
sensitivity analysis and no change was noticed. Lack of data
on previous comparison between sudomotor function tests
(Neuropad) and other diabetic neuropathy tests limit the
comparison of our results, which are particularly significant
in terms of savings when Neuropad is compared to SWME.
Moreover, we assumed that death rates were independent of
age since our analysis is not based on a cohort with diabetes,
but we use data from published economic evaluations on
diabetic peripheral neuropathy interventions and we assume
that the population modelled in this economic evaluation
would be any person of any age with any type of diabetes.
Additional analysis that could be undertaken to enhance
the robustness or completeness of the results would mainly
refer to the limitations named before (a full economic evalu-
ation with real patient data after randomised controlled trials
to assess the added value of Neuropad into daily practice;
more detailed data on SWME costs) or a local evaluation
of costs in an NHS hospital and community care home, to
make comparisons, where people with diabetes are screened
using Neuropad and more precise data on their particular
health status (other diabetes-related diagnoses) and costs of
care could be available. Furthermore, we could assess the
long-term cost-effectiveness of an SFT (Neuropad), alone
or together with the 10-g monofilament, using the United
Kingdom Prospective Diabetes Study (UKPDS) risk engine.
While national and international guidance on diabetic
foot and diabetic foot ulcers care are available, the evidence
base for much of routine clinical care is quite scarce [42].
There is currently no alternative test for diabetic neuropa-
thy that is suitable for patients to self-test. By encourag-
ing patients with diabetes to monitor their foot health using
Neuropad by performing the test at home, self-testing would
reduce the need for patients to attend a foot examination and
free up more time in the clinic or GP surgery. Longer term,
by reducing the incidence of ulcers and amputations, the
NHS would require fewer wound care specialists and sur-
geons, and these could be deployed elsewhere. It would also
save considerable costs on dressings, medications, antibiot-
ics and post-operative care. This paper provides with some
evidence on a cost-effective way to promote early detection
of diabetic neuropathy, combining a sudomotor function
Author's personal copy

B.Rodríguez-Sánchez et al.
1 3
self-test together with the current standard of care, the 10-g
SWME. Given the scope of economic evaluations to pro-
vide with evidence on the efficient allocation of healthcare
resources, our results suggest that using Neuropad and 10-g
SWME leads not only to cost savings, but also to health
gains. Therefore, the figures reported in this study could be
of help to ensure the sustainability of healthcare systems tak-
ing into account the high priority the British government has
given to diabetes and the substantial proportion of the NHS
budget, around 1%, devoted to foot care [43].
Acknowledgements This research did not receive any specific grant
from funding agencies in the public, commercial, or not-for-profit sec-
tors. We thank the contribution of John Simpson, Peter Altman and
Marta Trapero Bertrán, as well as the anonymous referees that have
reviewed different versions of this manuscript for their appreciated
input to this research paper.
Compliance with ethical standards
Conflicts of interest No potential conflicts of interest relevant to this
article were reported.
References
1. Diabetes UK.: Facts and Stats from https ://diabe tes-resou rcesp
roduc tion.s3euw est1.amazo naws.com/diabe tesst orage /migra tion/
pdf/Diabe tesUK _Facts _Stats _Oct16 .pdf (2016). Accessed 10 Apr
2017
2. Stratton, I.M., Adler, A.I., Neil, H.A.W., etal.: Association of
glycaemia with macrovascular and microvascular complications
of type 2 diabetes (UKPDS 35): prospective observational study.
BMJ 321, 405–412 (2000)
3. NIDDKD.: https ://www.niddk .nih.gov/healt h-infor matio n/diabe
tes/overv iew/preve nting -probl ems/nerve -damag e-diabe tic-neuro
pathi es (2018). Accessed 7 June 2018
4. Strachan, M.W.J., etal.: Cognitive function, dementia and type 2
diabetes mellitus in the elderly. Nat. Rev. Endocrinol. 7, 108–114
(2011)
5. Boulton, A.J.M.: Management of diabetic peripheral neuropathy.
Clin. Diabetes 23, 9–15 (2005)
6. Ziegler, D.: Diabetic peripheral neuropathy. In: Holt, R.I.G.,
Cockram, C.S., Flyvbjerg, A., etal. (eds.) Textbook of Diabetes,
4th edn. Wiley-Blackwell, Oxford (2010)
7. Kerr, M.: Footcare for people with diabetes: the economic case
for change. Insight Health Economics, NHS Diabetes (2012)
8. Public Health England.: Diabetes footcare activity profiles. In:
Using the Average Annual Number Of amputations Per Year from
2012 to 2015 (2016)
9. Public Health England.: Annual Report and Accounts 2014/15
(2015)
10. Kerr, M.: Diabetic Foot Care in England: An Economic Study.
Insight Health Economics, For Diabetes, London (2017)
11. Prashanth, R.J.V., Edmonds, M.E.: Early recognition of diabetic
peripheral neuropathy and the need for one-stop microvascular
assessment. Lancet Diabetes Endocrinol. 4(9), 723–725 (2016)
12. Gaede, P., Lund-Andersen, H., Parving, H.H., Pedersen, O.: Effect
of a multifactorial intervention on mortality in type 2 diabetes.
New Engl. J. Med. 358, 580–591 (2008)
13. Gaede, P., Vedel, P., Parving, H.H., Pedersen, O.: Intensified mul-
tifactorial intervention in patients with type 2 diabetes mellitus
and microalbuminuria: the Steno type 2 randomised study. Lancet
353(9153), 617–622 (1999)
14. Dros, J., Wewerinke, A., Bindels, P.J., van Weert, H.C.: Accuracy
of monofilament testing to diagnose peripheral neuropathy: a sys-
tematic review. Ann. Family Med. 7(6), 555–558 (2009)
15. Feng, Y., Schlösser, F.J., Sumpio, B.E.: The Semmes Weinstein
monofilament examination as a screening tool for diabetic periph-
eral neuropathy. J. Vasc. Surg. 50(3), 675–682 (2009)
16. National Institute for Health & Care Excellence.: Diabetic foot
problems. Prevention and management. NICE Clinical Guideline
No. 19. https ://www.nice.org.uk/guida nce/ng19/evide nce/full-
guida nce-pdf-15672 91554 3) (2016a). Accessed 5 June 2018
17. Sanz-Corbalán, I., Lázaro-Martínez, J.L., García-Morales, E.,
Molines-Barroso, R., Álvaro-Afonso, F., García-Álvarez, Y.:
Advantages of early diagnosis of diabetic neuropathy in the pre-
vention of diabetic foot ulcers. Diabetes Res. Clin. Pract. 146,
148–154 (2018)
18. Kim, C.H., Jeong, S.J., Mok, J.O., Lee, K.Y., Kim, S.R.: Impact
on quality of life in peoples with painful diabetic peripheral neu-
ropathy. Endocr. Abstr. 37, EP443 (2015)
19. Van Acker, K., Bouhassira, D., De Bacquer, D., Weiss, S., Mat-
thys, K., Raemen, H., Mathieu, C., Colin, I.M.: Prevalence and
impact on quality of life of peripheral neuropathy with or without
neuropathic pain in type 1 and type 2 diabetic patients attend-
ing hospital outpatients clinics. Diabetes Metab. 35(3), 206–213
(2009)
20. Quattrini, C., Jeziorska, M., Tavaloki, M., Begum, P., Boulton,
A.J., Mali, R.A.: The Neuropad test: a visual indicator test for
human diabetic neuropathy. Diabetologia 51(6), 1046–1050
(2008)
21. Papanas, N., Giassakis, G., Papatheodorou, K., etal.: Sensitivity
and specificity of a new indicator test (Neuropad) for the diagnosis
of peripheral neuropathy in type 2 diabetes patients: a comparison
with clinical examination and nerve conduction study. J. Diabetes
Complicat. 21(6), 353–358 (2007)
22. Papanas, N., Papatheodorou, K., Papazoglou, D., etal.: A compar-
ison of the new indicator test for sudomotor function (Neuropad)
with the vibration perception threshold and the clinical examina-
tion in the diagnosis of peripheral neuropathy in subjects with
type 2 diabetes. Exp. Clin. Endocrinol. Diabetes 116(2), 135–138
(2008)
23. Liatis, S., Marinou, K., Tentolouris, N., etal.: Usefulness of a new
indicator test for the diagnosis of peripheral and autonomic neu-
ropathy in patients with diabetes mellitus. Diabetic Med. 24(12),
1375–1380 (2007)
24. Zick, R., Schäper, T., Deeters, U.: Peripherie diabetische neuropa-
thie. Die schweibsekretion am fub messen. Klinikarzt 32, 192–194
(2003)
25. Crawford, F., McCowan, C., Dimitrov, B.D., Woodburn, J., Wylie,
G.H., Booth, E., Leese, G.P., Bekker, H.L., Kleijnen, J., Fahey,
T.: The risk of foot ulceration in people with diabetes screened in
community settings: findings from a cohort study. QJM 104(5),
403–410 (2011)
26. Edlin, R., McCabe, C., Hulme, C., Hall, P., Wright, J.: Building a
Markov cost effectiveness model in excel. In: Cost Effectiveness
Modelling for Health Technology Assessment, 1st edn, pp. 41–57.
Adis (2015)
27. Drummond, M.F., Sculpher, M.J., Claxton, K., Stoddart, G.L.,
Torrance, G.W.: Methods for the Economic Evaluation of Health
Care Programmes, 4th edn. Oxford University Press, Oxford
(2015)
28. Green, W., Taylor, M.: Cost-effectiveness analysis of d-Nav for
people with diabetes at high risk of neuropathic foot ulcers. Dia-
betes Ther. 7, 511–525 (2016)
Author's personal copy

Cost‑effectiveness analysis oftheNeuropad device asascreening tool forearly diabetic…
1 3
29. Ortegon, M.M., Redekop, W.K., Niessen, L.W.: Cost-effectiveness
of prevention and treatment of the diabetic foot. Diabetes Care 27,
901–907 (2004)
30. Ragnarson-Tennvall, G., Apelqvist, J.: Prevention of diabetes-
related foot ulcers and amputations: a cost-utility analysis based
on Markov model simulations. Diabetologia 44, 2077–2087
(2001)
31. National Institute for Health & Care Excellence.: NICE MTEP
briefing note—MT318 Neuropad for detecting diabetic foot neu-
ropathy. MT318 Neuropad—MTAC November 2016 (2016b)
32. Claxton, K., Paulden, M., Gravelle, H., Brouwer, W., Culyer, A.J.:
Discounting and decision making in the economic evaluation of
health-care technologies. Health Econ. 20, 2–15 (2011)
33. Brazier, J.: Valuing health states for use in cost-effectiveness
analysis. Pharmacoeconomics 26(9), 769–779 (2008)
34. Redekop, W.K., Stolk, E.A., Kok, E., Lovas, K., Kalo, Z., Bussch-
bach, J.J.V.: Diabetic foot ulcers and amputations: estimates of
health utility for use in cost-effectiveness analyses of new treat-
ments. Diabetes Metab 30, 549–556 (2004)
35. Claxton, K., Culyer, A.J., Schulper, M.J., McCabe, C., Briggs,
A., Akehurst, R., Brazier, J., Buxton, M.: Discounting and cost-
effectiveness in NICE—stepping back to sort out a confusion.
Health Econ. 15, 1–4 (2006)
36. Brigss, A., Claxton, K., Schulper, M.J.: Decision analytic model-
ling for the evaluation of health technologies. Oxford University
Press, Oxford (2006)
37. Claxton, K., Martin, S., Soares, M., Rice, N., Spackman, E.,
Hinde, S., Devlin, N., Smith, P.C., Schulper, M.: Methods for the
estimation of the NICE cost effectiveness threshold. CHE research
paper 81 (2013)
38. British Diabetic Association Good clinical practice guidelines for
care home residents with diabetes. A revision document prepared
by a Task and Finish Group of Diabetes UK. British Diabetic
Association for Diabetes UK (2010)
39. van Kollenburg, E.G., Lavrijsen, J.C., Verhagen, S.C., Zuidema,
S.U., Schalkwijk, A., Vissers, K.C.: Prevalence, causes and treat-
ment of neuropathic pain in Dutch nursing home residents: a ret-
rospective chart review. J. Am. Geriatr. Soc. 60(8), 1418–1425
(2012)
40. Davies, M., Brophy, S., Williams, R., Taylor, A.: The prevalence,
severity, and impact of painful diabetic peripheral neuropathy in
type 2 diabetes. Diabetes Care 29, 1518–1522 (2006)
41. Abbott, C.A., Malik, R.A., van Ross, E.R., Kulkarni, J., Boulton,
A.J.: Prevalence and characteristics of painful diabetic neuropa-
thy in a large community-based diabetic population in the U.K.
Diabetes Care 34(10), 2220–2224 (2011)
42. Jeffcoate, W.J., Vileikyte, L., Boyko, E.J., Armstrong, D.G., Boul-
ton, A.J.M.: Current challenges and opportunities in the preven-
tion and management of diabetic foot ulcers. Diabetes Care 41,
645–652 (2018)
43. Kerr, M., Barron, E., Chadwick, P., Evans, T., Kong, W.M., Ray-
man, G., Sutton-Smith, M., Todd, G., Young, B., Jeffcoate, W.J.:
The cost of diabetic foot ulcers and amputations to the National
Health Service in England. Diabet. Med. 8, 8–9 (2019). https ://
doi.org/10.1111/dme.13973
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Author's personal copy