This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Cost-Effectiveness of Interventions toPrevent and Control Diabetes Mellitus:A Systematic ReviewRUI LI, PHD
PING ZHANG, PHD
LAWRENCE E. BARKER, PHD
FARAH M. CHOWDHURY, MPH
XUANPING ZHANG, PHD
OBJECTIVE — To synthesize the cost-effectiveness (CE) of interventions to prevent andcontrol diabetes, its complications, and comorbidities.
RESEARCH DESIGN AND METHODS — We conducted a systematic review of liter-ature on the CE of diabetes interventions recommended by the American Diabetes Association(ADA) and published between January 1985 and May 2008. We categorized the strength ofevidence about the CE of an intervention as strong, supportive, or uncertain. CEs were classifiedas cost saving (more health benefit at a lower cost), very cost-effective (�$25,000 per life yeargained [LYG] or quality-adjusted life year [QALY]), cost-effective ($25,001 to $50,000 per LYGor QALY), marginally cost-effective ($50,001 to $100,000 per LYG or QALY), or not cost-effective (�$100,000 per LYG or QALY). The CE classification of an intervention was reportedseparately by country setting (U.S. or other developed countries) if CE varied by where theintervention was implemented. Costs were measured in 2007 U.S. dollars.
RESULTS — Fifty-six studies from 20 countries met the inclusion criteria. A large majority ofthe ADA recommended interventions are cost-effective. We found strong evidence to classify thefollowing interventions as cost saving or very cost-effective: (I) Cost saving— 1) ACE inhibitor(ACEI) therapy for intensive hypertension control compared with standard hypertension con-trol; 2) ACEI or angiotensin receptor blocker (ARB) therapy to prevent end-stage renal disease(ESRD) compared with no ACEI or ARB treatment; 3) early irbesartan therapy (at the microalbu-minuria stage) to prevent ESRD compared with later treatment (at the macroalbuminuria stage);4) comprehensive foot care to prevent ulcers compared with usual care; 5) multi-componentinterventions for diabetic risk factor control and early detection of complications compared withconventional insulin therapy for persons with type 1 diabetes; and 6) multi-component inter-ventions for diabetic risk factor control and early detection of complications compared withstandard glycemic control for persons with type 2 diabetes. (II) Very cost-effective— 1) intensivelifestyle interventions to prevent type 2 diabetes among persons with impaired glucose tolerancecompared with standard lifestyle recommendations; 2) universal opportunistic screening forundiagnosed type 2 diabetes in African Americans between 45 and 54 years old; 3) intensiveglycemic control as implemented in the UK Prospective Diabetes Study in persons with newlydiagnosed type 2 diabetes compared with conventional glycemic control; 4) statin therapy forsecondary prevention of cardiovascular disease compared with no statin therapy; 5) counselingand treatment for smoking cessation compared with no counseling and treatment; 6) annualscreening for diabetic retinopathy and ensuing treatment in persons with type 1 diabetes com-pared with no screening; 7) annual screening for diabetic retinopathy and ensuing treatment inpersons with type 2 diabetes compared with no screening; and 8) immediate vitrectomy to treatdiabetic retinopathy compared with deferred vitrectomy.
CONCLUSIONS — Many interventions intended to prevent/control diabetes are cost savingor very cost-effective and supported by strong evidence. Policy makers should consider givingthese interventions a higher priority.
Diabetes Care 33:1872–1894, 2010
The cost of diabetes in the U.S. in2007 was $174 billion (1). Many in-terventions can reduce the burden
of this disease. However, health care re-sources are limited; thus, interventionsfor diabetes prevention/control should beprioritized. We wanted to compare theeffectiveness and costs of various inter-ventions to find those that were the mosteffective for the least expense. Cost-effective analysis is a useful tool for thispurpose. Such analyses consist of compil-ing incremental cost-effectiveness ratios(ICERs), which are calculated as a ratio ofthe difference in costs to the difference ineffectiveness between the intervention be-ing evaluated and the comparisonintervention.
With the same health outcome indi-cator, ICERs of interventions are compa-rable. Therefore, these ICERs can make iteasier to decide how to allocate resources.Although many cost-effectiveness (CE)analyses of diabetes interventions havebeen published, their qualities and con-clusions vary. A systematic review, whichappraises individual studies and summa-rizes results, would aid policy makers andclinicians in prioritizing interventions toprevent or treat diabetes and itscomplications.
Few investigators have conductedsystematic reviews of the CE of diabetesinterventions (2–5). The systematicreview presented here, following theCochrane Collaboration’s protocol (6),includes all English language studiesavailable from 1985 to May 2008. Theinterventions included only those recom-mended by the 2008 American DiabetesAssociation (ADA) Standards of MedicalCare in Diabetes (7).
RESEARCH DESIGN ANDMETHODS
Study selection and protocols forreviewWe searched the Medical Literature Anal-ysis and Retrieval System Online (MED-LINE), Excerpta Medica (EMBASE),Cumulative Index to Nursing and AlliedHealth Literature (CINAHL), PsycINFO,Sociological Abstracts (Soc Abs), Web of
From the Division of Diabetes Translation, National Center for Chronic Disease Prevention and HealthPromotion, Centers for Disease Control and Prevention, Atlanta, Georgia.
Corresponding author: Rui Li, [email protected] findings and conclusions in this report are those of the authors and do not necessarily reflect the official
cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0/ for details.
R e v i e w s / C o m m e n t a r i e s / A D A S t a t e m e n t sR E V I E W A R T I C L E
1872 DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 care.diabetesjournals.org
Science (WOS), and Cochrane databasesto identify relevant studies. We created asearch strategy involving medical subjectheadings. The key words—and what eachindicated—were:
● Indicating diabetes: 26 key words indi-cating the disease of diabetes, such as“type 1 diabetes,” “type 2 diabetes,”“impaired glucose tolerance,” and “in-sulin resistance”;
● Indicating costs: (“cost or expendi-ture”) OR (“costs and cost analysis”) OR(“health care costs”) OR (“cost ofillness”);
● Indicating effectiveness: (“benefit”) OR(“life years”) OR (“quality-adjusted lifeyears”) OR (“disability adjusted lifeyears”);
● Indicating CE analysis: [(key words forcosts) AND (keywords for effective-ness)] OR (“cost-benefit analysis”) OR(“cost-effectiveness analysis”) OR(“cost-utility analysis”) OR (“economicevaluation”).
Database searches were based onmatches in all four keyword categories.Reference lists of all the included articleswere screened for additional citations,and Diabetes Care was reviewed manu-ally, issue by issue, as the journal was ex-pected to be highly relevant.
Criteria for inclusion in the reviewwere 1) original CE analysis; 2) interven-tion directed toward patients with type 1,type 2, or gestational diabetes mellitus(GDM) and recommended in the 2008ADA standards for medical care (7); 3)outcomes were measured as life yearsgained (LYGs) or quality-adjusted lifeyears gained (QALYs); and 4) publicationin the English language occurred betweenJanuary 1985 and May 2008 (2). To en-sure that only studies with acceptablequality were included, we limited theanalysis to studies considered good or ex-cellent according to a 13-item quality-assessment tool based on the BritishMedical Journal authors’ guide for eco-nomic studies (8).
To make ICERs comparable acrossthe studies, all costs are expressed as 2007U.S. dollars with adjustment from othercurrencies, as needed, using the FederalReserve Bank’s annual foreign exchangerates (9) and from other cost years usingthe Consumer Price Index (10). If a studydid not mention the year used in cost cal-culations, we assumed cost was as of oneyear before publication. ICERs were ex-pressed as dollars per QALY or dollars per
LYG and were rounded to the nearesthundred dollars per QALY or LYG.
Classification of cost-effectiveness ofinterventionsInterventions were classified based on thelevel of CE by convention as described inthe literature (2,11,12)—cost saving (anintervention generates a better health out-come and costs less than the comparisonintervention) or cost neutral (ICER � 0);very cost-effective (0 � ICER � $25,000per QALY or LYG); cost-effective($25,000 � ICER � $50,000 per QALY orLYG); marginally cost-effective ($50,000 �ICER � $100,000 per QALY or LYG); ornot cost-effective (�$100,000 per QALY orLYG)—and whether evidence for the inter-vention’s CE was strong, supportive, or un-certain as described below.
There were two grades of evidence in-cluded in the “strong” group. Grade 1 wasdefined as 1) CE of the intervention wasevaluated by two or more studies; 2)study quality was rated good or excellent;3) effectiveness of interventions based onwell-conducted, randomized clinicaltrials with adequate power and gen-eralizable results or meta-analysis or avalidated simulation model; 4) effec-tiveness of interventions rated as level A(clear evidence from well-conducted,generalizable, randomized controlled tri-als that were adequately powered; com-pelling nonexperimental evidence, i.e.,the all or none rule developed by theCentre for Evidence-Based Medicine atthe University of Oxford, U.K.) or levelB (supportive evidence from well-conducted cohort studies or supportiveevidence from a well-conducted case-control study) according to the 2008 ADAstandards of medical care (7); and 5) sim-ilar ICERs reported across the studies.Grade 2 was defined as the same as Grade1 except that the CE was based on onlyone study and the study was rated asexcellent.
We called the level of evidence “sup-portive” if only one study, rated lowerthan excellent, evaluated the CE of theintervention or if the effectiveness of theintervention was supported by eitherlevel C evidence (supportive evidencefrom poorly controlled or uncontrolledstudies, or conflicting evidence with theweight of evidence supporting the recom-mendation) or expert consensus (level E)in ADA recommendations (7). The term“uncertain” was used to describe inter-ventions with inconsistent evidenceabout CE across studies.
Reporting the results of thesystematic reviewWe reported the study results in twoways: 1) summarizing the key featuresand results for each included study; and2) synthesizing the CE of the interven-tions based on the classification criteriadescribed above. For the summary, wegrouped interventions based on their in-tended purposes: a) preventing type 2 di-abetes among high-risk persons; b)screening for undiagnosed type 2 diabetesand GDM; c) management of diabetes andrisk factors for complications; d) screen-ing for and early treatment of complica-tions; and e) treatment of complicationsand comorbidities. We considered caseswhere the same intervention was appliedto different populations or was comparedwith different interventions as differentspecific interventions and reported theICERs separately. This was because bothincremental costs and effectiveness of anintervention, and thus the ICERs, varied ifthe population and/or comparison groupdiffered. If the CE of an intervention wasevaluated from different study perspec-tives, we report the ICERs separately. Wepresented the ICERs in subgroups if theirICERs differed substantially from base-case analysis, and original studies re-ported the ICERs this way. If the studyreported the ICERs only for populationsubgroups, we provided a range and,when available, trend of the ICERs. Fi-nally, if a study used both LYGs andQALYs as study outcome measures, wereported the ICER in both costs per LYGand QALY.
In reporting the synthesized results,we applied the following rules: 1) Weused the median ICER to represent the CEof an intervention if the intervention wasevaluated by more than one study. 2) Wereported the ICERs from the longer ana-lytical time horizon if the interventionwas evaluated from both short- and long-term perspectives. This was appropriatesince many of the benefits of most diabe-tes prevention and control interventionswould come from preventing diabeticcomplications, which occur later in life. 3)We chose the health care system as ourprimary study perspective for the purposeof cross-study and cross-interventioncomparisons. This study perspective in-cluded all the medical costs incurred nomatter who paid. 4) If the ICERs of anintervention differed substantially be-tween the U.S. and other developed coun-tries (mainly European countries,Australia, and Canada), we reported the
Li and Associates
care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 1873
summary results separately by labelingthe ICER for the U.S. or for the othercountries. 5) If the trial on which the CEof an intervention was based was con-ducted in a mixed population with type 1or type 2 diabetes, we assumed the CEwas the same for both types of diabetes.
RESULTS — The search yielded 9,461abstracts. After reviewing the abstractsand subsequent reference tracking, wenarrowed the focus to 197 possible origi-nal CE studies. Further review of the fulltext resulted in 56 CE studies that met ourinclusion criteria. Figure 1 depicts thedata abstraction process.
Table 1 shows the detailed descrip-tion of the CE studies that we includedaccording to intervention type (13–70).
We first grouped similar interventions to-gether, then arranged them chronologi-cally and by the first author’s last name.Some studies that evaluated multiple in-terventions appear in more than one cat-egory. The information used to describeeach study included the intervention be-ing evaluated; comparison intervention,population, and country setting; datasources for the effectiveness of the inter-vention; study methods; quality of thestudy; analytical time horizon; discountrate (a rate that is used to convert futurecosts and benefits into their present val-ues); and ICER.
Thirty-nine of the 56 studies took along-term analytical time horizon, such as20–30 years or lifetime. Nearly all of thestudies with the long-term horizon used
simulation modeling. Only one study wasconducted in a developing country (Thai-land) (57). There were 48 excellent stud-ies and 8 good studies. Only three studiestook perspectives other than the healthcare system.
The interventions evaluated in theseCE studies covered a wide range: lifestyleand medication therapy to prevent type 2diabetes among high-risk individuals(eight studies); screening for undiag-nosed type 2 diabetes or GDM (threestudies); intensive glycemic control (12studies); self-monitoring of blood glucose(one study); intensive hypertension con-trol (four studies); statin therapy for cho-lesterol control (five studies); smokingcessation (one study); diabetic health ed-ucation program (two studies); diabetes
Figure 1—Selection of cost-effectiveness studies for systematic review of interventions to prevent and control diabetes.
Cost-effectiveness of diabetes interventions
1874 DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 care.diabetesjournals.org
Tab
le1—
Des
crip
tion
ofth
eco
st-e
ffec
tive
ness
stud
ies
for
diab
etes
inte
rven
tion
s*
Sour
ce/s
tudy
qual
ity†
/cou
ntry
Stud
ypo
pula
tion
Inte
rven
tion
‡C
ompa
riso
nE
ffec
tive
ness
data
Met
hodo
logy
�/an
alyt
ical
hori
zon/
disc
ount
rate
Cos
t-ef
fect
iven
ess
rati
os(2
007
U.S
.$)
Pre
ven
tin
gty
pe
2d
iabe
tes
amon
gh
igh
-ris
kin
div
idu
als
Sega
let
al.1
998
(59)
§A
ustr
alia
Seri
ousl
yob
ese
orse
riou
sly
obes
ew
ith
IGT
Inte
nsiv
edi
etan
ded
ucat
ion
Stan
dard
care
Lite
ratu
rere
view
25ye
ars
5%C
ost
savi
ng
Ove
rwei
ght
orob
ese
IGT
orN
GT
and
IGT
Gro
uped
ucat
ion
inw
orkp
lace
ondi
etan
dph
ysic
alac
tivi
tyfo
rm
en
Stan
dard
care
Cos
tsa
ving
Hig
h-ri
skad
ults
IGT
orN
GT
and
IGT
Gen
eral
prac
titi
oner
advi
ceon
heal
thy
lifes
tyle
Stan
dard
care
$1,0
00–$
2,50
0/LY
G
Ove
rwei
ght
adul
tsin
gene
ralp
opul
atio
nC
omm
unit
y-su
ppor
ted
med
iaca
mpa
ign
onob
esit
y/se
dent
ary
lifes
tyle
No
cam
paig
nC
ost
savi
ng
Wom
enw
ith
GD
Mhi
stor
y�
NG
Tor
IGT
Inte
nsiv
edi
etan
dbe
havi
oral
mod
ifica
tion
Stan
dard
care
$1,3
00–$
2,50
0/LY
G
DPP
2003
(66)
U.S
.IG
TIn
tens
ive
lifes
tyle
mod
ifica
tion
Stan
dard
advi
ceon
lifes
tyle
DPP
Mul
tice
nter
RC
T(n
�3,
234)
3ye
ars
0%$3
2,90
0/Q
ALY
;ifi
n10
-per
son
grou
p,$1
1,10
0/Q
ALY
IGT
Met
form
inSt
anda
rdad
vice
onlif
esty
le$1
34,0
00/Q
ALY
;ifm
etfo
rmin
cost
redu
ced
50%
,$76
,500
/QA
LYIG
TIn
tens
ive
lifes
tyle
mod
ifica
tion
‡‡St
anda
rdad
vice
onlif
esty
le$6
9,40
0/Q
ALY
;ifi
n10
-per
son
grou
p,$3
6,00
0/Q
ALY
IGT
Met
form
in‡‡
Stan
dard
advi
ceon
lifes
tyle
$133
,400
/QA
LYC
aro
etal
.200
4(1
5)C
anad
aIG
TIn
tens
ive
lifes
tyle
mod
ifica
tion
No
inte
rven
tion
DPP
(n�
3,23
4),
FDPS
(n�
52)
10ye
ars
5%$7
00/L
YGIG
TM
etfo
rmin
No
inte
rven
tion
Cos
tsa
ving
inLY
Gan
dQ
ALY
Palm
eret
al.2
004
(50)
Aus
tral
ia,
IGT
Inte
nsiv
elif
esty
lem
odifi
cati
onSt
anda
rdad
vice
onlif
esty
leD
PPM
ulti
cent
erR
CT
(n�
3,23
4)Li
feti
me
5%ex
cept
U.K
.:co
st5%
,C
ost
savi
ngex
cept
U.K
.;U
.K.:
$8,3
00/L
YGFr
ance
,G
erm
any,
Swit
zerl
and,
U.K
.
IGT
Met
form
inSt
anda
rdad
vice
onlif
esty
leef
fect
iven
ess,
1.5%
Cos
tsa
ving
,exc
ept
UK
;UK
:$6,
500/
LYG
Edd
yet
al.2
005
(25)
U.S
.IG
TIn
tens
ive
lifes
tyle
mod
ifica
tion
‡‡N
oin
terv
enti
onD
PPM
ulti
cent
erR
CT
(n�
3,23
4)30
year
s3%
$84,
700/
QA
LY;i
n10
-per
son
grou
p,$1
6,00
0/Q
ALY
IGT
Inte
nsiv
elif
esty
lem
odifi
cati
on#
No
inte
rven
tion
$192
,600
/QA
LY;i
n10
-per
son
grou
p,$3
6,40
0/Q
ALY
IGT
Met
form
in‡‡
No
inte
rven
tion
$47,
900/
QA
LYH
erm
anet
al.2
005
(34)
U.S
.IG
TIn
tens
ive
lifes
tyle
mod
ifica
tion
Stan
dard
advi
ceon
lifes
tyle
DPP
Mul
tice
nter
RC
T(n
�3,
234)
Life
tim
e3%
$1,5
00/Q
ALY
;in
10-p
erso
ngr
oup,
cost
savi
ngIn
tens
ive
lifes
tyle
mod
ifica
tion
‡‡St
anda
rdad
vice
onlif
esty
le$1
1,80
0/Q
ALY
Met
form
inSt
anda
rdad
vice
onlif
esty
le$4
2,00
0/Q
ALY
Gen
eric
Met
form
in‡‡
Stan
dard
advi
ceon
lifes
tyle
$2,4
00/Q
ALY
$40,
200/
QA
LYLi
ndgr
enet
al.
2007
(41)
Swed
en
IGT
Age
60ye
ars
BMI
�25
kg/m
2,F
PG�
6.1
mm
ol/l
Inte
nsiv
elif
esty
lein
terv
enti
on(6
year
s)‡‡
Gen
eral
lifes
tyle
advi
ceFD
PS(n
�52
)Li
feti
me
3%C
ost
savi
ngno
tco
nsid
erin
gco
stof
exte
nded
life;
$2,6
00/Q
ALY
incl
udin
gco
stof
exte
nded
life
Li and Associates
care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 1875
Hoe
ger
etal
.200
7(3
6)U
.S.
U.S
.pop
ulat
ion
age
45–
74ye
ars,
over
wei
ght
and
obes
e(B
MI
�25
kg/m
2)
Gro
ups
Scre
enin
gfo
rIG
Tan
dIF
PG,D
PPlif
esty
lein
terv
enti
onw
ith
IGT
�IF
PG
No
scre
enin
gan
dno
lifes
tyle
inte
rven
tion
DPP
(n�
3,23
4)Li
feti
me
3%$1
0,60
0/Q
ALY
;in
grou
pse
ttin
gs,
cost
savi
ng
Scre
enin
gfo
rIG
Tan
dIF
PG,D
PPlif
esty
lein
terv
enti
onw
ith
IFPG
orIG
T�
IFPG
No
scre
enin
gan
dlif
esty
lein
terv
enti
on$1
2,30
0/Q
ALY
;in
grou
pse
ttin
gs,
$344
/QA
LY
Scre
enin
gfo
rIG
Tan
dIF
PG,D
PPlif
esty
lein
terv
enti
onw
ith
IGT
�IF
PG
Scre
enin
gfo
rIG
Tan
dIF
PG,f
ollo
win
gD
PPlif
esty
lein
terv
enti
onw
ith
IFPG
,IG
T,o
rIF
PG�
IGT
$13,
100/
QA
LY
Scre
enin
gan
dm
etfo
rmin
trea
tmen
tw
ith
IGT
�IF
PGN
osc
reen
ing
and
trea
tmen
t$2
6,60
0/Q
ALY
Scre
enin
gan
dm
etfo
rmin
trea
tmen
tw
ith
IGT
,IFP
G,o
rIG
T�
IFPG
No
scre
enin
gan
dtr
eatm
ent
$26,
000/
QA
LY
Scre
enin
gfo
ru
nd
iagn
osed
typ
e2
dia
bete
san
dge
stat
ion
ald
iabe
tes
Cen
ters
for
Dis
ease
Con
trol
and
Prev
enti
on19
98(1
6)U
.S
U.S
.pop
ulat
ion
25ye
ars
and
olde
rO
ne-t
ime
Opp
ortu
nist
icsc
reen
ing
for
undi
agno
sed
type
2di
abet
esst
arti
ngat
age
25ye
ars,
then
trea
tmen
t(u
nive
rsal
scre
enin
g)
No
scre
enin
gan
dtr
eatm
ent
unti
lclin
ical
diag
nosi
sof
type
2di
abet
es
Life
tim
e3%
$374
,900
/LYG
or$8
9,80
0/Q
ALY
;in
crea
sing
wit
hag
e(a
ge�
25ye
ars)
trea
tmen
t(u
nive
rsal
scre
enin
g).
$57,
100/
LYG
or$2
1,40
0/Q
ALY
(age
25–3
4ye
ars)
$103
,200
/LYG
or$2
9,70
0/Q
ALY
(age
35–4
4ye
ars)
293,
900/
LYG
or$7
0,10
0/Q
ALY
(age
45–5
4ye
ars)
$1m
illio
n/LY
Gor
$185
,000
/QA
LY(a
ge55
–64
year
s)$9
28,0
00/Q
ALY
(age
�65
)A
fric
anA
mer
ican
s:ag
e25
–34
year
s$3
,500
/LYG
or$1
,300
/QA
LYag
e35
–44
year
s$1
0,20
0/LY
Gor
$3,1
00/Q
ALY
age
45–5
4ye
ars
$95,
400/
LYG
or$1
9,60
0/Q
ALY
age
55–6
4ye
ars
$764
,100
/LYG
or$1
12,6
00/Q
ALY
age
�65
year
s$2
mill
ion/
LYG
or$5
00,0
00/Q
ALY
Hoe
rger
etal
.200
4(3
5)U
.S.
Pers
ons
wit
hhy
pert
ensi
onT
arge
ted
scre
enin
gfo
run
diag
nose
ddi
abet
esam
ong
pers
ons
wit
hhy
pert
ensi
on
No
scre
enin
gor
trea
tmen
tun
tilc
linic
aldi
agno
sis
ofty
pe2
diab
etes
Life
tim
e3%
$46,
800–
$130
,500
/QA
LYde
crea
sing
wit
hag
e$7
0,50
0/Q
ALY
for
age
45ye
ars
U.S
.pop
ulat
ion
One
-tim
eop
port
unis
tic
scre
enin
g,th
entr
eatm
ent
(uni
vers
alsc
reen
ing)
No
scre
enin
gor
trea
tmen
tun
tilc
linic
aldi
agno
sis
ofty
pe2
diab
etes
$72,
200–
$189
,100
/QA
LYde
crea
sing
wit
hag
e$1
83,5
00/Q
ALY
for
age
45ye
ars
U.S
.pop
ulat
ion
One
-tim
eop
port
unis
tic
scre
enin
g,th
entr
eatm
ent
(uni
vers
alsc
reen
ing)
Tar
gete
dsc
reen
ing,
then
trea
tmen
t$2
15,6
00–$
699,
800/
QA
LYin
crea
sing
wit
hag
e
Nic
olso
net
al.2
005
(44)
U.S
.30
-yea
r-ol
dpr
egna
ntw
omen
betw
een
Sequ
enti
alm
etho
d(5
0-g
GC
T�
100-
gG
TT
)‡‡
No
scre
enin
g75
-gG
TT
Afe
wun
iden
tifie
dR
CT
s�
1ye
ar**
0%C
ost
savi
ng
24–2
8w
eeks
’ge
stat
ion
100-
gG
TT
‡‡N
osc
reen
ing
or75
-gG
TT
met
hod
Cos
tsa
ving
100-
gG
TT
‡‡Se
quen
tial
met
hod
$35,
200/
QA
LYfo
rm
ater
nal
outc
omes
,$9,
000/
QA
LYfo
rne
onat
alou
tcom
es
Cost-effectiveness of diabetes interventions
1876 DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 care.diabetesjournals.org
Inte
nsi
vegl
ycem
icco
ntr
olD
CC
T19
96(6
5)U
.S.
Typ
e1
diab
etes
Inte
nsiv
egl
ycem
icco
ntro
lth
roug
hin
sulin
man
agem
ent,
self-
mon
itor
ing,
and
outp
atie
ntvi
sits
.The
goal
was
toac
hiev
eA
1Cle
vela
sno
rmal
aspo
ssib
le(6
%)
Con
vent
iona
lthe
rapy
(les
sin
tens
ive)
DC
CT
Mul
tice
nter
RC
T(n
�1,
441)
Life
tim
e3%
$47,
600/
life
year
gain
ed,$
50,8
00/
QA
LY
Palm
eret
al.2
000
(46)
Swit
zerl
and
Typ
e1
diab
etes
Inte
nsiv
ein
sulin
ther
apy
Con
vent
iona
lins
ulin
ther
apy
Lite
ratu
rere
view
Life
tim
e3%
,5%
,6%
Rep
orte
dre
sult
sat
3%in
the
tabl
e
$46,
600/
LYG
Scuf
fham
etal
.20
03(5
8)U
.K.
Typ
e1
diab
etes
Con
tino
ussu
bcut
aneo
usin
sulin
inte
rven
tion
for
pers
ons
usin
gin
sulin
pum
p
Mul
tipl
eda
ilyin
sulin
inje
ctio
ns1
syst
emat
icre
view
1m
eta-
anal
ysis
8ye
ars
6%$1
0,20
0/Q
ALY
Roz
eet
al.2
005
(56)
U.K
.T
ype
1dia
bete
sC
onti
nuou
ssu
bcut
aneo
usin
sulin
infu
sion
Mul
tipl
eda
ilyin
sulin
inje
ctio
nsD
CC
T(n
�1,
441)
mai
nly
met
a-an
alys
is
60ye
ars
3%$1
8,50
0/Q
ALY
Eas
tman
etal
.199
7(2
4)U
.S.
New
lydi
agno
sed
type
2di
abet
esIn
tens
ive
trea
tmen
tta
rget
ing
mai
nten
ance
ofA
1Cle
vela
t7.
2%
Stan
dard
anti
diab
etic
trea
tmen
tta
rget
ing
A1C
leve
lat
10%
DC
CT
(n�
1,44
1)Li
feti
me
3%$1
7,40
0/Q
ALY
;sen
siti
veto
age
atdi
abet
eson
set;
CE
R�
33,0
00fo
rag
e�
50ye
ars;
$371
,700
/QA
LYfo
rag
e70
–80
year
sG
ray
etal
.200
0(3
0)U
.K.
Typ
e2
diab
etes
Inte
nsiv
em
anag
emen
tw
ith
insu
linor
sulfo
nylu
rea
aim
ing
atFP
G�
6m
mol
/l
Con
vent
iona
lman
agem
ent
(mai
nly
thro
ugh
diet
)ai
min
gat
FPG
�15
mm
ol/l
UK
PDS
Mul
tice
nter
RC
T(n
�5,
120)
10ye
ars*
*6%
Cos
tsa
ving
intr
ial;
$1,1
00/e
vent
-fre
eye
arga
ined
incl
inic
sett
ing
Wak
eet
al.2
000
(70)
Japa
nT
ype
2di
abet
esIn
tens
ive
insu
linth
erap
yth
roug
hm
ulti
ple
insu
linin
ject
ions
A1C
�7%
Con
vent
iona
lins
ulin
inje
ctio
nth
erap
yK
umam
oto
stud
yR
CT
(n�
110)
10ye
ars*
*3%
Cos
tsa
ving
Cla
rke
etal
.200
1(1
8)U
.K.
New
lydi
agno
sed
type
2di
abet
esO
verw
eigh
tIn
tens
ive
bloo
dgl
ucos
eco
ntro
lw
ith
met
form
inai
min
gat
FPG
�6
mm
ol/l
Con
vent
iona
ltre
atm
ent
prim
arily
wit
hdi
etU
KPD
S(n
�5,
120)
Med
ian
10.7
year
s**
6%C
ost
savi
ng
Cen
ters
for
Dis
ease
Con
trol
and
Prev
enti
on20
02(1
7)U
.S.
New
lydi
agno
sed
type
2di
abet
esIn
tens
ive
glyc
emic
cont
rolw
ith
insu
linor
sulfo
nylu
rea
aim
ing
atFP
Gof
6m
mol
/l
Con
vent
iona
lglu
cose
cont
rol(
mai
nly
diet
)U
KPD
S(n
�5,
120)
Life
tim
e3%
$62,
000/
QA
LY;i
ncre
asin
gra
pidl
yw
ith
age
atdi
agno
sis:
$14,
400/
QA
LYfo
rag
e25
–34
year
s;$2
7,50
0–$5
6,00
0/Q
ALY
for
age
35–5
4ye
ars;
�$1
00,0
00–$
3.1
mill
ion
for
age
55–9
4ye
ars
Cos
tsa
ving
unde
rU
KPD
Sco
stsc
enar
io(n
oca
sem
anag
emen
tco
st,
muc
hle
ssse
lf-te
stin
g,sl
ight
lyfe
wer
phys
icia
nvi
sits
)bu
tus
ing
U.S
.uni
tco
stC
lark
eet
al.2
005
(19)
U.K
.N
ewly
diag
nose
dty
pe2
diab
etes
requ
irin
gin
sulin
Inte
nsiv
egl
ycem
icco
ntro
lwit
hin
sulin
orsu
lfony
lure
aat
FPG
�6
mm
ol/l
Con
vent
iona
lglu
cose
cont
rolt
hera
py(m
ainl
ydi
et)
UK
PDS
(n�
5,12
0)Li
feti
me
3.5%
$3,4
00/Q
ALY
New
lydi
agno
sed
type
2di
abet
esO
verw
eigh
tIn
tens
ive
glyc
emic
cont
rolw
ith
met
form
inC
onve
ntio
nalg
luco
seco
ntro
lthe
rapy
(mai
nly
diet
)
Cos
tsa
ving
Li and Associates
care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 1877
Edd
yet
al.2
005
(25)
U.S
.N
ewly
diag
nose
dty
pe2
diab
etes
Inte
nsiv
eD
PPlif
esty
lew
ith
FPG
�12
5m
mol
/lT
arge
t:A
1Cle
velo
f7%
‡‡
Die
tary
advi
ceD
PP(n
�3,
234)
30ye
ars
3%$3
3,10
0/Q
ALY
Alm
bran
det
al.
2000
(13)
Swed
en
Typ
e2
diab
etes
wit
hac
ute
MI
Insu
lin-g
luco
sein
fusi
onfo
rat
leas
t24
h,th
ensu
bcut
aneo
usm
ulti
dose
insu
linfo
r�
3m
onth
s
Stan
dard
anti
diab
etic
ther
apy
DIG
AM
Ist
udy,
RC
T1-
year
inte
rven
tion
,4-
year
follo
w-u
p(n
�62
0)
5ye
ars*
*3%
$8,7
00/L
YG,$
12,4
00/Q
ALY
Self
-mon
itor
ing
bloo
dgl
uco
seT
unis
2008
(67)
U.S
.T
ype
2di
abet
estr
eate
dw
ith
oral
agen
tsin
aSM
BG1
tim
e/da
y40
-yea
rho
rizo
npu
blic
paye
rN
oSM
BGK
aise
rPe
rman
ente
long
itud
inal
stud
y40
year
s3%
$8,2
00/Q
ALY
;52.
6%pr
obab
ility
less
than
$50,
000/
QA
LYla
rge
HM
OSM
BG3
tim
es/d
ay40
-yea
rho
rizo
nN
oSM
BGof
coho
rtof
“new
anti
diab
etic
drug
$6,9
00/Q
ALY
;60.
7%pr
obab
ility
less
than
$50,
000/
QA
LYSM
BG1
tim
e/da
y5-
year
hori
zon
10-y
ear
hori
zon
No
SMBG
user
s”$2
4,20
0/Q
ALY
$9,7
00/Q
ALY
SMBG
3ti
mes
/day
5-ye
arho
rizo
n10
-yea
rho
rizo
nN
oSM
BG$3
0,30
0/Q
ALY
$540
/QA
LYIn
ten
sive
hyp
erte
nsi
onco
ntr
olU
KPD
S19
98(6
8)U
.K.
Typ
e2
diab
etes
Hyp
erte
nsio
nT
ight
cont
rolo
fhyp
erte
nsio
n,BP
�15
0/�
80m
mH
g,A
CE
inhi
bito
r,�
-blo
cker
,and
othe
rag
ents
Less
tigh
tco
ntro
lofB
P(m
mH
g),I
niti
ally
�20
0/10
5,La
ter
180/
105
UK
PDS
(n�
5,12
0)Li
feti
me
6%C
ost
savi
ngin
tria
l;$9
60/y
ear
free
from
occu
rren
ceof
diab
etes
endp
oint
incl
inic
Elli
otet
al.2
000
(26)
U.S
.T
ype
2di
abet
esH
yper
tens
ion,
init
ially
free
ofC
VD
orE
SRD
Red
ucti
onof
BPto
130/
85m
mH
gM
edic
atio
nsno
tm
enti
oned
Red
ucti
onof
BPto
140/
90m
mH
gM
eta-
anal
ysis
ofda
tafr
omep
idem
iolo
gica
l
Life
tim
e3%
Star
tof
trea
tmen
tst
udie
san
dcl
inic
alA
ge50
year
str
ials
$1,2
00/L
YGA
ge60
year
sC
ost
savi
ngA
ge70
year
sC
ost
savi
ngC
ente
rsfo
rD
isea
seC
ontr
olan
dPr
even
tion
2002
(17)
U.S
.
Typ
e2
diab
etes
Hyp
erte
nsio
nIn
tens
ified
hype
rten
sion
cont
rol
AC
Ein
hibi
tor
�-b
lock
erA
vera
geBP
144/
82m
mH
g
Mod
erat
ehy
pert
ensi
onco
ntro
l,A
vera
geBP
154/
86m
mH
g
UK
PDS
(n�
5,12
0)Li
feti
me
3%C
ost
savi
ng
Cla
rke
etal
.200
5(1
9)U
.K.
Typ
e2
diab
etes
Hyp
erte
nsio
nT
ight
BPco
ntro
lBP
�15
0/85
mm
Hg,
AC
Ein
hibi
tor
(cap
topr
il)or
�-b
lock
er(a
teno
lol)
Less
tigh
tco
ntro
lofB
P(m
mH
g),I
niti
al�
200/
105,
Late
r�
180/
105
UK
PDS
(n�
5,12
0)Li
feti
me
3.5%
$200
/QA
LY
Ch
oles
tero
lco
ntr
olH
erm
anet
al.1
999
(33)
U.S
.T
ype
2di
abet
esD
yslip
idem
ia,P
revi
ous
MI
oran
gina
Sim
vast
atin
Plac
ebo
4Sst
udy,
Dou
ble-
blin
dra
ndom
ized
,pl
aceb
o-co
ntro
lled,
mul
tice
nter
,m
ulti
coun
try
tria
l(n
�4,
444)
5ye
ars*
*3%
for
cost
,0%
for
bene
fitC
ost
savi
ng
Jons
son
etal
.199
9(3
9)E
urop
ean
coun
trie
s
Typ
e2
diab
etes
Dys
lipid
emia
,Pre
viou
sM
Ior
angi
na
Sim
vast
atin
Plac
ebo
4Sst
udy
(n�
4,44
4)Li
feti
me
3%C
S-$9
,400
/LYG
indi
ffer
ent
coun
trie
s,M
edia
n:$2
,800
/LYG
Cost-effectiveness of diabetes interventions
1878 DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 care.diabetesjournals.org
Gro
ver
etal
.200
0(3
1)C
anad
aT
ype
2di
abet
esD
yslip
idem
iaC
VD
hist
ory,
Men
and
wom
en60
year
sol
d
Sim
vast
atin
Plac
ebo
4Sst
udy
(n�
4,44
4)C
AR
E(n
�4,
159)
Life
tim
e5%
$6,1
00–$
12,3
00/L
YGIn
crea
sing
wit
hpr
etre
atm
ent
ofLD
Lch
oles
tero
lle
vel
Typ
e2
diab
etes
Dys
lipid
emia
,No
CV
Dhi
stor
y
Sim
vast
atin
Plac
ebo
Men Pr
etre
atm
ent
LDL
chol
este
roll
evel
:5.
46m
mol
/l(2
11m
g/dl
)$6
,100
–$15
,000
/LYG
3.5
mm
ol/l
(135
mg/
dl)
$10,
700–
$23,
000/
LYG
Wom
enPr
etre
atm
ent
LDL
chol
este
roll
evel
:5.
46m
mol
/l3.
5m
mol
/l$1
5,30
0–$2
7,60
0/LY
G$3
6,80
0–$6
1,30
0/LY
GC
ente
rsfo
rD
isea
seC
ontr
olan
dPr
even
tion
2002
(17)
U.S
.
Typ
e2
diab
etes
Dys
lipid
emia
,No
CV
Dhi
stor
y
Prav
asta
tin
Plac
ebo
Wes
tSc
otla
ndC
oron
ary
Prev
enti
onSt
udy
(n�
6,59
5m
en)
Life
tim
e3%
U-s
hape
for
age,
$77,
800/
QA
LY
Rai
kou
etal
.200
7(5
4)U
.K.I
rela
ndT
ype
2di
abet
es,N
oC
VD
hist
ory,
No
elev
ated
LDL
chol
este
roll
evel
�1
CV
Dri
skfa
ctor
:re
tino
path
y,m
icro
albu
min
uria
orm
acro
albu
min
uria
,cu
rren
tsm
okin
g,or
hype
rten
sion
Ato
rvas
tati
nPl
aceb
oC
AR
DS,
Ran
dom
ized
,co
ntro
lled,
mul
tice
nter
tria
l94
%w
hite
(n�
2,83
8)
Life
tim
e3.
5%$2
,800
/LYG
,$3,
500/
QA
LYU
sing
UK
PDS
risk
engi
neLo
wri
sk:
$11,
300/
QA
LY;M
ediu
mri
sk:
$4,7
00/Q
ALY
;Hig
hri
sk:$
2,20
0/Q
ALY
Smok
ing
cess
atio
nE
arns
haw
etal
.20
02(2
3)N
ewly
diag
nose
dty
pe2
diab
etes
Smok
ing
cess
atio
n,St
anda
rdan
tidi
abet
icca
reSt
anda
rdan
tidi
abet
icca
reLi
feti
me
3%
Uni
ted
Stat
esSm
oker
sA
ged
25–8
4ye
ars
�$2
5,00
0/Q
ALY
Age
d85
–94
year
s$8
9,80
0/Q
ALY
Ed
uca
tion
alp
rogr
amSh
eare
ret
al.2
004
(61)
§G
erm
any
Typ
e1
diab
etes
Stru
ctur
edtr
eatm
ent
and
teac
hing
prog
ram
:edu
cati
onal
cour
seof
trai
ning
tose
lf-m
anag
edi
abet
esan
den
joy
diet
ary
free
dom
Usu
alca
re(d
aily
insu
linin
ject
ion)
Ros
iglit
azon
etr
ial
CO
DE
2st
udy
ofpr
eval
ence
ofco
mpl
icat
ions
,not
anR
CT
Life
tim
e6%
Cos
tsa
ving
Goz
zoli
etal
.200
1(2
9)Sw
itze
rlan
dT
ype
2St
anda
rdan
tidi
abet
icca
repl
used
ucat
iona
lpro
gram
,Sel
f-m
onit
orin
g,R
ecom
men
dati
ons
ondi
etan
dex
erci
se,S
elf-
man
agem
ent
ofdi
abet
esan
dco
mpl
icat
ions
,Gen
eral
heal
thed
ucat
ion
Stan
dard
anti
diab
etic
care
Lite
ratu
rere
view
(qua
lity)
Life
tim
e3%
$4,0
00/L
YG
Li and Associates
care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 1879
Dia
bete
sd
isea
sem
anag
emen
tM
ason
etal
.200
5(4
3)E
ngla
ndT
ype
2di
abet
esH
yper
tens
ion
Polic
yto
impl
emen
tcl
inic
sle
dby
spec
ialis
tnu
rses
totr
eat
and
cont
rolh
yper
tens
ion
thro
ugh
cons
ulta
tion
,med
icat
ion
revi
ew,c
ondi
tion
asse
ssm
ent,
and
lifes
tyle
advi
ce
Usu
alca
reSP
LIN
TR
CT
(n�
1,40
7)U
KPD
S(n
�5,
120)
Life
tim
e5%
$4,8
00/Q
ALY
Dia
gnos
eddi
abet
esD
yslip
idem
iaPo
licy
toim
plem
ent
clin
ics
led
bysp
ecia
list
nurs
esto
trea
tan
dco
ntro
lhyp
erlip
idem
iaby
usua
lcar
e
Usu
alca
re$2
3,60
0/Q
ALY
Gilm
eret
al.2
007
(27)
San
Die
goC
ount
y,C
alifo
rnia
Dia
bete
s48
%La
tino
sC
ultu
rally
sens
itiv
eca
sem
anag
emen
tan
dse
lf-m
anag
emen
ttr
aini
ngpr
ogra
mle
dby
bilin
gual
/bic
ultu
ral
med
ical
assi
stan
tan
dre
gist
ered
diet
itia
nst
eppe
d-ca
reph
arm
acol
ogic
man
agem
ent
ofgl
ucos
ean
dlip
idle
vels
and
hype
rten
sion
‡‡‡
Stan
dard
care
Proj
ect
Dul
ceO
bser
vati
onal
coho
rtst
udy
wit
hco
ntro
lsA
vera
gefo
llow
-up,
289
days
(n�
3,89
3)
40ye
ars
3%$9
,400
/LYG
or$1
2,00
0/Q
ALY
for
unin
sure
d;10
0%pr
obab
ility
tobe
less
than
$50,
000
and
$100
,000
/Q
ALY
,res
pect
ivel
y
$22,
400/
LYG
or$2
9,10
0/Q
ALY
for
pati
ents
inC
ount
yM
edic
alSe
rvic
es;
92%
or98
%pr
obab
ility
tobe
cost
-ef
fect
ive
ifw
illin
gnes
sto
pay
was
$50,
000
or$1
00,0
00/Q
ALY
,re
spec
tive
ly$4
2,60
0/LY
Gor
$53,
120/
QA
LYfo
rpa
tien
tsin
Med
i-C
al;
57%
or81
%pr
obab
ility
tobe
cost
-ef
fect
ive
ifw
illin
gnes
sto
pay
was
$50,
000
and
$100
,000
/QA
LY,
resp
ecti
vely
$68,
400/
LYG
or$8
2,30
0/Q
ALY
for
pati
ents
wit
hco
mm
erci
alin
sura
nce;
31%
and
62%
prob
abili
tyto
beco
st-
effe
ctiv
eif
will
ingn
ess
topa
yw
as$5
0,00
0an
d$1
00,0
00/Q
ALY
,re
spec
tive
lyP
reve
nti
ng
dia
beti
cco
mp
lica
tion
sE
yeco
mp
lica
tion
sJa
vitt
etal
.199
4(3
7)§
U.S
.N
ewly
diag
nose
dty
pe2
diab
etes
8st
rate
gies
for
eye
scre
enin
gw
ith
dila
tion
:Scr
eeni
ngev
ery
1,2,
3,or
4ye
ars
and
No
scre
enin
gC
ross
-sec
tion
alan
dlo
ngit
udin
alst
udie
s
Life
tim
e5%
All
8st
rate
gies
wer
eco
stsa
ving
Mor
efr
eque
ntfo
llow
-up
scre
enin
gfo
rdi
abet
espa
tien
tsw
ith
back
grou
ndre
tino
path
y††
Cost-effectiveness of diabetes interventions
1880 DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 care.diabetesjournals.org
Javi
ttet
al.1
996
(38)
U.S
.N
ewly
diag
nose
dty
pe1
and
type
2di
abet
esA
nnua
leye
scre
enin
gw
ith
dila
tion
for
allp
atie
nts
wit
hE
yesc
reen
ing
in60
%of
diab
etes
pati
ents
Cro
ss-s
ecti
onal
and
long
itud
inal
Life
tim
e5%
$3,8
00/p
erso
n-ye
arof
sigh
tsa
ved,
$6,9
00/Q
ALY
Typ
e1
diab
etes
diab
etes
but
nore
tino
path
yst
udie
s$4
,300
/QA
LYT
ype
2di
abet
esE
xam
inat
ion
ever
y6
mon
ths
for
thos
ew
ith
reti
nopa
thy
$6,9
00/Q
ALY
Palm
eret
al.2
000
(46)
Swit
zerl
and
Typ
e1
diab
etes
Ann
uale
yesc
reen
ing
and
trea
tmen
t,C
onve
ntio
nal
insu
linth
erap
y
Con
vent
iona
lins
ulin
ther
apy
Lite
ratu
rere
view
Life
tim
e3%
Cos
tsa
ving
Vija
net
al.2
000
(69)
U.S
.T
ype
2E
yesc
reen
ing
for
diab
etes
pati
ents
ever
y5
year
sSu
bseq
uent
annu
alsc
reen
ing
for
thos
ew
ith
back
grou
ndre
tino
path
y
No
scre
enin
gE
pide
mio
logi
cal
stud
ies
Life
tim
e3%
$23,
500/
QA
LY
Eye
scre
enin
gfo
rdi
abet
espa
tien
tsev
ery
3ye
ars
Subs
eque
ntan
nual
scre
enin
gfo
rth
ose
wit
hba
ckgr
ound
reti
nopa
thy
No
scre
enin
g$2
7,00
0/Q
ALY
Eye
scre
enin
gfo
rdi
abet
espa
tien
tsev
ery
2ye
ars
Subs
eque
ntan
nual
scre
enin
gfo
rth
ose
wit
hba
ckgr
ound
reti
nopa
thy
No
scre
enin
g$3
0,70
0/Q
ALY
Eye
scre
enin
gan
nual
lyfo
rdi
abet
espa
tien
tsSu
bseq
uent
annu
alsc
reen
ing
for
thos
ew
ith
back
grou
ndre
tino
path
y
No
scre
enin
g$3
9,50
0/Q
ALY
Eye
scre
enin
gfo
rdi
abet
espa
tien
tsev
ery
3ye
ars
5-ye
arin
terv
als
$32,
800/
QA
LY
Eye
scre
enin
gfo
rdi
abet
espa
tien
tsev
ery
2ye
ars
3-ye
arin
terv
als
$54,
000/
QA
LY
Ann
uale
yesc
reen
ing
for
diab
etes
pati
ents
2-ye
arin
terv
als
$116
,800
/QA
LY
Mab
erle
yet
al.
2003
(42)
Wes
tern
Jam
esBa
y,V
icto
ria,
Brit
ish
Col
umbi
a,C
anad
a
Typ
e1
diab
etes
and
Typ
e2
diab
etes
Scre
enin
gus
ing
digi
talc
amer
aIm
med
iate
asse
ssm
ent
ofqu
alit
yor
elec
tron
ical
lytr
ansf
erre
dto
are
mot
ere
adin
gce
nter
Ret
ina
spec
ialis
tsvi
sit
Moo
seFa
ctor
yev
ery
6m
onth
sto
exam
ine
peop
lew
ith
diab
etes
,and
pati
ents
inou
tlyi
ngco
mm
unit
ies
are
flow
nto
Moo
seFa
ctor
y,C
anad
a
10ye
ars
5%C
ost
savi
ng
Foo
tu
lcer
sT
ennv
alet
al.2
001
(64)
Swed
enT
ype
1di
abet
esan
dT
ype
2di
abet
esO
ptim
alpr
even
tion
offo
otul
cer
incl
udin
gfo
otin
spec
tion
,U
sual
care
Clin
ical
and
epid
emio
logi
cal
5ye
ars*
*0%
Hig
hri
sk:
appr
opri
ate
foot
wea
r,da
taPr
evio
usfo
otul
cer
Prev
ious
ampu
tati
ontr
eatm
ent,
and
educ
atio
nC
ost
savi
ng
Mod
erat
eri
sk:
Neu
ropa
thy,
PVD
,and
/or
foot
defo
rmit
yC
ost
savi
ng
Li and Associates
care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 1881
Low
risk
:N
osp
ecifi
cri
skfa
ctor
�$1
00,0
00/Q
ALY
Ort
egon
etal
.N
ewly
diag
nose
dty
pe2
Inte
nsiv
egl
ycem
icco
ntro
lSt
anda
rdca
reU
KPD
S(n
�5,
120)
Life
tim
e3%
$44,
900/
QA
LY20
04§
(45)
The
Net
herl
ands
diab
etes
Foot
ulce
rO
ptim
alfo
otca
reLi
tera
ture
revi
ewon
tria
lsan
dA
ssum
ing
10%
redu
ctio
nof
foot
lesi
on,$
308,
300/
QA
LYep
idem
iolo
gica
lst
udie
sA
ssum
ing
90%
redu
ctio
nof
foot
lesi
on,$
17,0
00/Q
ALY
Inte
nsiv
egl
ycem
icco
ntro
lplu
sop
tim
alfo
otca
reSt
anda
rdca
reA
ssum
ing
10%
redu
ctio
nof
foot
lesi
on,$
34,4
00/Q
ALY
Ass
umin
g90
%re
duct
ion
offo
otle
sion
,$11
,010
/QA
LYE
nd
-sta
gere
nal
dis
ease
Borc
h-Jo
hnse
net
al.
1993
(14)
§G
erm
any
Typ
e1
diab
etes
Ann
uals
cree
ning
for
mic
roal
bum
inur
iaat
5ye
ars
afte
rdi
abet
eson
set,
AC
EI
trea
tmen
t
Tre
atm
ent
ofm
acro
albu
min
uria
Dan
ish
coho
rt(n
�2,
890)
30ye
ars
6%C
ost
savi
ng
Kib
erd
etal
.199
6(4
0)§
Can
ada
Typ
e1
diab
etes
Scre
enin
gfo
rm
icro
albu
min
uria
AC
EI
trea
tmen
tT
reat
men
tof
hype
rten
sion
and/
orm
acro
prot
einu
ria
Clin
ical
tria
lLi
feti
me
5%$5
8,40
0/Q
ALY
Palm
eret
al.2
000
(46)
Swit
zerl
and
Typ
e1
diab
etes
Hig
hto
talc
hole
ster
olle
vel
Hig
hsy
stol
icBP
Mic
roal
bum
inur
iam
onit
orin
g,A
CE
trea
tmen
t,C
onve
ntio
nal
insu
linth
erap
y
Con
vent
iona
lins
ulin
ther
apy
Lite
ratu
rere
view
Life
tim
e3%
Cos
tsa
ving
Don
get
al.2
004
(22)
U.S
.T
ype
1di
abet
esA
CE
Itr
eatm
ent
star
ting
at1
year
afte
rdi
agno
sis
Ann
uals
cree
ning
for
mic
roal
bum
inur
iaA
CE
trea
tmen
t
DC
CT
(n�
1,44
1)Li
feti
me
3%$3
8,00
0/Q
ALY
,Inc
reas
edw
ith
low
erin
gA
1Cle
vel;
atA
1Cle
vel
9%,�
25,0
00/Q
ALY
Sakt
hong
etal
.20
01(5
7)§
Tha
iland
Typ
e2
diab
etes
Mic
roal
bum
inur
iabu
tno
rmal
BP
AC
Ein
hibi
tors
Plac
ebo
7-ye
arR
CT
inIs
rael
(n�
94)
25ye
ars
8%C
ost
savi
ng
Souc
het
etal
.200
3(6
2)Fr
ance
Typ
e2
diab
etes
Nep
hrop
athy
Losa
rtan
Plac
ebo
RE
NA
AL
stud
yM
ulti
cent
erin
tern
atio
nalt
rial
(n�
1,51
3)
4ye
ars*
*C
ost
disc
ount
edat
8%Be
nefit
sno
tdi
scou
nted
Cos
tsa
ving
Szuc
set
al.2
004
(63)
§Sw
itze
rlan
d
Typ
e2
diab
etes
Nep
hrop
athy
Losa
rtan
Plac
ebo
RE
NA
AL
stud
yM
ulti
cent
erin
tern
atio
nalt
rial
(n�
1,51
3)
3.5
year
s**
0%C
ost
savi
ng
Palm
eret
al.2
003
(47)
Belg
ium
,Fr
ance
Typ
e2
diab
etes
Mac
roal
bum
inur
iaH
yper
tens
ion
Irbe
sart
anSt
anda
rdth
erap
yfo
rhy
pert
ensi
onID
NT
stud
yM
ulti
cent
er,
doub
le-b
lind
plac
ebo
cont
rolle
dtr
ial(
n�
1,71
5)
Life
tim
e3%
Cos
tsa
ving
Palm
eret
al.2
004
(49)
U.K
.T
ype
2di
abet
esH
yper
tens
ion
Nep
hrop
athy
Irbe
sart
anSt
anda
rdth
erap
yfo
rhy
pert
ensi
onID
NT
stud
y(n
�1,
715)
10ye
ars
6%fo
rco
sts
1.5%
for
bene
fits
Cos
tsa
ving
Palm
eret
al.2
005
(51)
Spai
nT
ype
2di
abet
esM
icro
albu
min
uria
Hyp
erte
nsio
n
Irbe
sart
anSt
anda
rdth
erap
yfo
rhy
pert
ensi
on,N
oA
CE
I,A
IIR
A,o
r�
-blo
cker
s
IDN
Tst
udy
(n�
1,71
5)IR
MA
-2tr
ial
Ran
dom
ized
cont
rolle
dst
udy
(n�
582)
25ye
ars
3%C
ost
savi
ng
Palm
eret
al.2
007
(52)
Hun
gary
Typ
e2
diab
etes
Mic
roal
bum
inur
iaA
ddin
gir
besa
rtan
Plac
ebo
�st
anda
rdth
erap
yfo
rhy
pert
ensi
onID
NT
stud
y(n
�1,
715)
IRM
A-2
tria
l(n
�58
2)
25ye
ars
5%C
ost
savi
ng
Cost-effectiveness of diabetes interventions
1882 DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 care.diabetesjournals.org
Palm
eret
al.2
004
(48)
U.S
.T
ype
2di
abet
esH
yper
tens
ion
Irbe
sart
anat
stag
eof
mic
roal
bum
inur
iaSt
anda
rdth
erap
yfo
rhy
pert
ensi
onID
NT
stud
y(n
�1,
715)
25ye
ars
3%C
ost
savi
ng
Mic
roal
bum
inur
iaIr
besa
rtan
atst
age
ofm
acro
albu
min
uria
Stan
dard
ther
apy
for
hype
rten
sion
Cos
tsa
ving
Irbe
sart
anat
stag
eof
mic
roal
bum
inur
iaIr
besa
rtan
atst
age
ofm
acro
albu
min
uria
Cos
tsa
ving
Palm
eret
al.2
007
(53)
U.K
.T
ype
2di
abet
esH
yper
tens
ion
Irbe
sart
anat
stag
eof
mic
roal
bum
inur
iaSt
anda
rdth
erap
yfo
rhy
pert
ensi
onID
NT
stud
y(n
�1,
715)
IRM
A-2
tria
l(n
�58
2)
25ye
ars
3.5%
Cos
tsa
ving
Mic
roal
bum
inur
iaIr
besa
rtan
atst
age
ofm
acro
albu
min
uria
Stan
dard
ther
apy
for
hype
rten
sion
Cos
tsa
ving
Irbe
sart
anat
stag
eof
mic
roal
bum
inur
iaIr
besa
rtan
atst
age
ofm
acro
albu
min
uria
Cos
tsa
ving
Coy
leet
al.2
007
(21)
Can
ada
Typ
e2
diab
etes
Hyp
erte
nsio
nM
acro
neph
ropa
thy
orM
icro
neph
ropa
thy
Irbe
sart
anad
ded
atst
age
ofm
icro
albu
min
uria
Con
vent
iona
ltre
atm
ent
for
diab
etes
and
hype
rten
sion
,No
AC
EI
orA
IIR
As
IDN
Tst
udy
(n�
1.71
5)IR
MA
-2tr
ial(
n�
582)
Life
tim
e5%
Cos
tsa
ving
Irbe
sart
anad
ded
atst
age
ofov
ert
neph
ropa
thy
Con
vent
iona
ltre
atm
ent
for
diab
etes
and
hype
rten
sion
Cos
tsa
ving
Irbe
sart
anad
ded
atst
age
ofm
icro
albu
min
uria
Irbe
sart
anad
ded
atst
age
ofov
ert
neph
ropa
thy
Cos
tsa
ving
Gol
anet
al.1
999
(28)
U.S
.N
ewly
diag
nose
dty
pe2
diab
etes
Tre
atpa
tien
tsw
ith
new
diag
nosi
sw
ith
AC
EI
Scre
enin
gfo
rm
acro
albu
min
uria
and
trea
tmen
tw
ith
AC
EI
U.S
.-C
anad
aC
olla
bora
tive
stud
yfo
rty
pe1
diab
etes
,RC
T(n
�20
7)2
RC
Tfo
rty
pe2
diab
etes
inIs
rael
(n�
94an
d15
6,re
spec
tive
ly)
Life
tim
e3%
Cos
tsa
ving
Scre
enin
gfo
rm
icro
albu
min
uria
and
trea
tmen
tw
ith
AC
EI
Scre
enin
gfo
rm
acro
albu
min
uria
and
trea
tmen
tw
ith
AC
EI
Cos
tsa
ving
Tre
atpa
tien
tsw
ith
new
diag
nosi
sw
ith
AC
EI
Scre
enin
gfo
rm
icro
albu
min
uria
and
trea
tmen
tw
ith
AC
EI
$10,
900/
QA
LY
Cla
rke
etal
.200
0(2
0)C
anad
aT
ype
1di
abet
esPr
ovin
ceor
terr
itor
ypa
ying
for
AC
EI
Pay
from
out-
of-p
ocke
tC
olla
bora
tive
obse
rvat
iona
lst
udy
usin
gad
min
istr
ativ
eda
taba
se(N
�8.
4m
illio
n)
21ye
ars
5%C
ost
savi
ngC
ompl
ianc
era
tean
dco
stof
AC
EI
affe
cted
ICE
Rgr
eatl
y
Ros
enet
al.2
005
(55)
US
Med
icar
epo
pula
tion
Typ
e1
and
type
2di
abet
es
Med
icar
efu
ll-pa
ymen
tfo
rA
CE
IM
edic
are
payi
ngfo
rA
CE
IPa
yfr
omou
t-of
-poc
ket
Cur
rent
Med
icar
eM
oder
niza
tion
Act
HO
PET
rial
Mul
tina
tion
alR
CT
Life
tim
e3%
Cos
tsa
ving
ifA
CE
Ius
ein
crea
sed
byat
leas
t7.
2%If
use
incr
ease
dby
2.9%
,�$2
0,00
0/Q
ALY
Cos
tsa
ving
ifA
CE
Ius
ein
crea
sed
byat
leas
t6.
2%If
use
incr
ease
dby
2.2%
,�$2
0,00
0/Q
ALY
Li and Associates
care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 1883
Com
pre
hen
sive
inte
rven
tion
sPa
lmer
etal
.200
0(4
6)Sw
itze
rlan
dT
ype
1di
abet
esC
�A
CE
Ith
erap
y�
eye
scre
enin
gan
dtr
eatm
ent
(EYE
)C
onve
ntio
nalg
lyce
mic
cont
rol(
C)
Lite
ratu
rere
view
Life
tim
e3%
Cos
tsa
ving
Inte
nsiv
ein
sulin
ther
apy
(I)
�A
CE
Ith
erap
yI
$46,
500/
LYG
I�
EYE
I$5
0,60
0/LY
GI
�A
CE
Ith
erap
y�
EYE
I$4
9,80
0/LY
GG
ozzo
liet
al.2
001
(29)
Swit
zerl
and
Typ
e2
diab
etes
Add
eded
ucat
ion
prog
ram
,ne
phro
path
ysc
reen
ing,
and
AC
EI
ther
apy
tost
anda
rdan
tidi
abet
icca
re
Stan
dard
anti
diab
etic
care
Lite
ratu
rere
view
Life
tim
e0%
,3%
Cos
tsa
ving
Add
eded
ucat
ion
prog
ram
,ne
phro
path
ysc
reen
ing,
AC
EI
ther
apy,
and
reti
nopa
thy
scre
enin
gan
dla
ser
ther
apy
tost
anda
rdan
tidi
abet
icca
re
Stan
dard
anti
diab
etic
care
Cos
tsa
ving
Mul
tifa
ctor
iali
nter
vent
ion
incl
uded
educ
atio
nalp
rogr
am,
scre
enin
gfo
rne
phro
path
yan
dre
tino
path
y,co
ntro
lofC
VD
risk
fact
ors,
earl
ydi
agno
sis
and
trea
tmen
tof
com
plic
atio
ns,
and
heal
thed
ucat
ion
Stan
dard
anti
diab
etic
care
Cos
tsa
ving
Tre
atm
ent
ofd
iabe
tes-
rela
ted
com
pli
cati
ons
Ret
inop
ath
ySh
arm
aet
al.2
001
(60)
U.S
.D
iabe
tic
reti
nopa
thy
Hea
lth
mai
nten
ance
orga
niza
tion
Imm
edia
tevi
trec
tom
yfo
rm
anag
emen
tof
vitr
eous
hem
orrh
age
seco
ndar
yto
diab
etic
reti
nopa
thy
Def
erra
lofv
itre
ctom
yD
RV
SLi
feti
me
6%$2
,900
/QA
LY
Foo
tu
lcer
Hab
ache
ret
al.
2007
(32)
Aus
tria
New
lydi
agno
sed
diab
etic
foot
ulce
rIn
tens
ified
trea
tmen
tby
inte
rnat
iona
lcon
sens
uson
diab
etic
foot
care
Stan
dard
trea
tmen
tR
etro
spec
tive
stud
yof
pati
ent
reco
rds
on11
9co
nsec
utiv
eul
cera
tion
sin
86pa
tien
tsat
tert
iary
outp
atie
ntcl
inic
spec
ializ
ing
intr
eatm
ent
ofdi
abet
icfo
otul
cers
15ye
ars
0–8%
Cos
tsa
ving
4S,S
cand
inav
ian
Sim
vast
atin
Surv
ival
Stud
y;A
CE
I,an
giot
ensi
nco
nver
ting
enzy
me
inhi
bito
rs;A
HT
,art
eria
lhyp
erte
nsio
n;A
IIR
A,a
ngio
tens
inII
rece
ptor
anta
goni
sts;
BP,
bloo
dpr
essu
re;C
,con
vent
iona
lgly
cem
icco
ntro
l;C
AD
,cor
onar
yar
tery
dise
ase;
CA
RD
S,C
olla
bora
tiveA
torv
asta
tinD
iabe
tesS
tudy
;CA
RE
,Cho
lest
erol
and
Recu
rren
tEve
nts;
CD
C,C
ente
rsfo
rDis
ease
Con
trol
and
Prev
entio
n;C
OD
E2
�th
ecos
tofd
iabe
test
ype2
inEu
rope
;CO
RE
,Cen
terf
orO
utco
mes
Rese
arch
;CV
D,c
ardi
ovas
cula
rdis
ease
;DA
IS,D
iabe
tesA
ther
oscl
eros
isIn
terv
entio
nSt
udy;
DC
CT
,Dia
bete
sCon
trol
and
Com
plic
atio
nsTr
ial;
DIG
AM
I,D
iabe
tesM
ellit
usIn
sulin
Glu
cose
Infu
sion
inA
cute
Myo
card
ialI
nfar
ctio
n;D
iGE
M,
diab
etes
glyc
emic
educ
atio
nan
dm
onito
ring;
DPN
,dia
betic
perip
hera
lneu
ropa
thy;
DPP
,dia
bete
spre
vent
ion
prog
ram
;DR
VS,
Dia
betic
Retin
opat
hyV
itrec
tom
ySt
udy;
DT
TP,
diab
etes
trea
tmen
tand
teac
hing
prog
ram
;EY
E,s
cree
ning
for
retin
opat
hyan
den
suin
gtr
eatm
ent;
FDPS
,Fin
ish
Dia
bete
sPr
even
tion
Stud
y;FP
G,f
astin
gpl
asm
agl
ucos
e;H
MO
,Hea
lthM
aint
enan
ceO
rgan
izat
ion;
HO
PE,H
eart
Out
com
ePr
even
tion
Eval
uatio
n;I,
inte
nsiv
egl
ycem
icco
ntro
l;IC
ER
,in
crem
enta
lcos
teffe
ctiv
enes
srat
io;I
DN
T,I
rbes
arta
nTy
peII
Dia
betic
Nep
hrop
athy
Tria
l;IF
PG,i
mpa
ired
fast
ing
plas
mag
luco
se;I
GT
,im
paire
dgl
ucos
etol
eran
ce;I
MPA
CT
,Im
prov
ing
Moo
d-Pr
omot
ing
Acc
esst
oC
olla
bora
tiveT
reat
men
t;K
OR
A,C
oope
rativ
eRe
sear
chin
the
Regi
onof
Aug
sbur
g;M
I,m
yoca
rdia
linf
arct
ion;
NG
T,n
orm
algl
ucos
eto
lera
nce;
NID
DM
,Non
-Ins
ulin
Dep
ende
ntD
iabe
tesM
ellit
us;O
GT
T,o
ralg
luco
seto
lera
nce
test
;PH
N,p
osth
erpe
ticne
ural
gia;
PRO
acti
ve,P
ROsp
ectiv
epi
oglit
Azo
neC
linic
alTr
iali
nm
acro
Vas
cula
rEv
ents
;PR
OPH
ET
,Pro
spec
tive
Popu
latio
nH
ealth
Even
tTab
ulat
ion;
PVD
,per
iphe
ralv
ascu
lar
dise
ase;
RC
T,r
ando
miz
edcl
inic
altr
ial;
RE
NA
AL,
Redu
ctio
nof
Endp
oint
sin
NID
DM
with
the
Ang
iote
nsin
IIA
ntag
onis
tLos
arta
n;R
OSS
O,R
etrO
lect
ive
Stud
ySe
lf-M
onito
ring
ofBl
ood
Glu
cose
and
Out
com
e;R
PG,r
ando
mpl
asm
agl
ucos
e;SM
BG
,sel
f-mon
itorin
gbl
ood
gluc
ose;
SPE
CT
,sin
gle
prot
onem
issi
onco
mpu
ted
tom
ogra
phy;
SPLI
NT
,spe
cial
istn
urse
-led
inte
rven
tion
totr
eata
ndco
ntro
lhyp
erte
nsio
nan
dhy
perli
pide
mia
indi
abet
es;Q
ALY
,qua
lity
adju
sted
lifey
ear;
VA
-HIT
,VA
-HD
LIn
terv
entio
nTr
ial.
*The
stud
iesw
ereo
rder
edby
grou
ping
sim
ilari
nter
vent
ions
toge
ther
,the
nfo
llow
the
year
and
alph
abet
ical
orde
roft
hefir
stau
thor
’sla
stna
me;
the
num
bers
inth
epa
rent
hesi
sare
the
refe
renc
enu
mbe
r.†T
hest
udy
was
rate
das
“exc
elle
nt”q
ualit
yun
less
othe
rwis
ein
dica
ted.
§The
stud
yw
asra
ted
as“g
ood”
qual
ity.�
Thes
tudy
isba
sed
onsi
mul
atio
nm
odel
ing
unle
ssot
herw
isei
ndic
ated
.**W
ithin
tria
lorw
ithin
epid
emio
logi
cals
tudy
.‡Th
estu
dyw
asdo
nefr
omth
eper
spec
tiveo
fthe
heal
thsy
stem
unle
ssot
herw
ise
indi
cate
d.‡‡
The
stud
yw
asdo
nefr
omth
eso
ciet
alpe
rspe
ctiv
e.#T
hest
udy
done
from
the
pers
pect
ive
ofth
ehe
alth
plan
.††T
hest
udy
was
done
from
the
fede
ralb
udge
tper
spec
tive.
†††T
hird
part
ypa
yerp
ersp
ectiv
e.
Cost-effectiveness of diabetes interventions
1884 DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 care.diabetesjournals.org
disease management program (two stud-ies); screening to prevent diabetic reti-nopathy (five studies); optimal foot careto prevent foot ulcer and amputation (twostudies); ACE inhibitor (ACEI) or angio-tensin receptor blocker (ARB) therapy toprevent diabetic end-stage renal diseases(ESRD) (15 studies); comprehensive in-terventions using a combination of sev-eral of the above secondary preventioninterventions (two studies); and interven-tions treating diabetic retinopathy andfoot ulcers (two studies).
The classification of the interventionsbased on their level of CE and strength ofevidence is presented in Table 2. For eachintervention, we also described the num-ber of studies that evaluated the CE of thisintervention, its comparison interven-tion, and the study population in whichthe intervention was implemented. Wereported the median and range of the IC-ERs across the studies.
Twenty-six interventions were classi-fied as supported by strong evidence con-cerning their CE (Table 2). Among these,six interventions were cost saving, eightwere very cost-effective, six were cost-effective, two were marginally cost-effective, and four were not cost-effective.These interventions consisted of primaryprevention, screening for undiagnosedtype 2 diabetes, diabetic risk factor con-trol, early prevention of diabetes compli-cations, and treatment of diabetescomplications.
The six cost-saving interventions withstrong evidence were 1) ACEI therapy forintensive hypertension control, as in theUK Prospective Diabetes Study (UKPDS),in persons with type 2 diabetes comparedwith standard hypertension control; 2)ACEI or ARB therapy to prevent ESRD fortype 2 diabetes compared with no ACEIor ARB therapy; 3) early irbesartan ther-apy at the stage of microalbuminuria toprevent ESRD in people with type 2 dia-betes compared with treatment at thestage of macroalbuminuria; 4) compre-hensive foot care to prevent ulcers inmixed population with either type 1 ortype 2 diabetes compared with usual care;5) multi-component interventions for di-abetic risk factor control and early detec-tion of complications compared withconventional insulin therapy for personswith type 1 diabetes; and 6) multi-component interventions for diabetic riskfactor control and early detection of com-plications compared with standard glyce-mic control for persons with type 2diabetes.
Of the eight very cost-effective inter-ventions with strong evidence, six werefor persons with type 2 diabetes, one forpersons with type 1 diabetes, and one fora mixed population with type 1 or type 2diabetes. Interventions for type 2 diabetesincluded: 1) primary prevention throughintensive lifestyle modification; 2) univer-sal opportunistic screening for un-diagnosed type 2 diabetes in AfricanAmericans between 45 and 54 years old;3) intensive glycemic control as imple-mented in UKPDS; 4) statin therapy forsecondary prevention of cardiovasculardisease; 5) smoking cessation; and 6) an-nual screening for diabetic retinopathyand early treatment of it. The interventionfor type 1 diabetes was annual screeningfor diabetic retinopathy and treating thepositive cases. The intervention for mixedpopulation of type 1 and type 2 diabeteswas immediate vitrectomy to treat dia-betic retinopathy compared with deferralof vitrectomy.
The six cost-effective interventionswith strong evidence were 1) one-timeopportunistic targeted screening for un-diagnosed type 2 diabetes in hypertensivepersons aged 45 years and older com-pared with no screening; 2) intensive in-sulin treatment for persons with type 1diabetes compared with conventional gly-cemic control; 3) UKPDS-like intensiveglycemic control applied to the U.S.health care system among adults youngerthan age 54 years with type 2 diabetescompared with conventional glycemiccontrol; 4) intensive glycemic control by aDiabetes Prevention Program (DPP) typeof intensive lifestyle intervention in per-sons with newly diagnosed type 2 diabe-tes compared with conventional glycemiccontrol; 5) statin therapy for primary pre-vention of cardiovascular disease in per-sons with type 2 diabetes compared withno statin therapy; 6) multi-component in-terventions including insulin therapy,ACEI therapy, and screening for retinop-athy in persons with type 1 diabetes com-pared with intensive insulin therapy.
The two marginally cost-effective in-terventions with strong evidence were 1)intensive glycemic control for all U.S. res-idents with type 2 diabetes diagnosed atage 25 years and older compared withusual care; and 2) screening for diabeticretinopathy every two years comparedwith screening every three years in per-sons with type 2 diabetes.
The four interventions with strongevidence of not being cost-effective were1) one-time universal opportunistic
screening for undiagnosed type 2 diabetesamong those aged 45 years and oldercompared with no screening; 2) universalscreening for type 2 diabetes comparedwith targeted screening; 3) intensive gly-cemic control in the U.S. setting for pa-tients diagnosed with diabetes at olderages (55–94 years of age) compared withusual care; and 4) annual screening forretinopathy compared with screening ev-ery two years. All these studies were fortype 2 diabetes.
There were 18 specific interventionsfor which their CEs were based only on“supportive” evidence. Among them, 15were each supported by one CE study, 13were supported by level C or level E evi-dence, and five were supported by level Aor B evidence as defined in the 2008 ADAstandards of medical care in diabetes (7).For those interventions with level A or Bevidence, the CE of each intervention wasevaluated by one study with a quality ofbeing “good.”
In terms of the level of the CE, 10 ofthe 18 specific interventions based on“supportive” evidence were cost-saving,including 1) screening using the sequen-tial method (50-g glucose challenge testfollowed by 100-g glucose tolerance test[GTT]) for GDM in 30-year-old pregnantwomen between 24–28 weeks’ gestationcompared with no screening; 2) screeningfor GDM using the 100-g GTT methodcompared with no screening; 3) the se-quential method compared with 75-gGTT screening for GDM; 4) 100-g GTTcompared with 75-g GTT screening forGDM; 5) diabetes self-management edu-cation for persons with type 1 diabetescompared with no education; 6) full-reimbursement policy for ACEI for pa-tients with type 1 diabetes compared withpatients paying out-of-pocket; 7) full-reimbursement policy for ACEI for pa-tients with type 2 diabetes compared withpatients paying out-of-pocket; 8) screen-ing using a mobile camera at a remote areaand processing data in a reading centercompared with a retina specialist’s visit ina mixed population of type 1 and type 2diabetes; 9) screening for diabetic ne-phropathy and ensuing ACEI or ARBtherapy in persons with type 1 diabetescompared with no screening; and 10) in-tensified foot ulcer treatment in a mixedpopulation with type 1 or type 2 diabetescompared with standard treatment.
Seven of the 18 specific interventionswere very cost-effective: 1) primary pre-vention of type 2 diabetes in women withGDM history through intensive lifestyle
Li and Associates
care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 1885
Tab
le2—
Sum
mar
yof
the
cost
-eff
ecti
vene
ssst
udie
sby
inte
rven
tion
*
Inte
rven
tion
Com
pari
son
Inte
rven
tion
popu
lati
onN
umbe
rof
stud
ies
Leve
lofr
ecom
men
dati
onby
AD
AM
edia
nof
the
cost
-ef
fect
iven
ess
rati
osR
ange
ofth
eco
st-e
ffec
tive
ness
rati
os
Stro
ng
evid
ence
Cos
tsa
vin
gA
CE
Ith
erap
yfo
rin
tens
ive
hype
rten
sion
cont
rol
Stan
dard
hype
rten
sion
cont
rol
Typ
e2
4B
Cos
tsa
ving
Cos
tsa
ving
-$1,
200/
LYG
$230
/Q
ALY
Add
itio
nof
AC
EI
orA
RB
ther
apy
topr
even
tE
SRD
No
AC
EI
orA
RB
ther
apy
Typ
e2
7A
Cos
tsa
ving
Cos
tsa
ving
Irbe
sart
anth
erap
yat
the
stag
eof
mic
roal
bum
inur
ia
Irbe
star
tan
ther
apy
atth
est
age
ofm
acro
albu
min
uria
Typ
e2
3A
Cos
tsa
ving
Cos
tsa
ving
Com
preh
ensi
vefo
otca
reto
prev
ent
ulce
r†U
sual
care
Mix
edpo
pula
tion
ofty
pe1
and
type
21
BC
ost
savi
ngC
ost
savi
ng
Mul
ti-c
ompo
nent
Con
vent
iona
lins
ulin
Typ
e1
1A
:AC
EI
trea
tmen
tC
ost
savi
ngC
ost
savi
ngin
terv
enti
ons
(con
vent
iona
lins
ulin
cont
rol,
AC
EI
trea
tmen
t,ey
esc
reen
ing,
and
trea
tmen
t)
cont
rol
B:ey
esc
reen
ing
and
ensu
ing
trea
tmen
t
Mul
ti-c
ompo
nent
Stan
dard
anti
diab
etic
Typ
e2
1B:
educ
atio
nC
ost
savi
ngC
ost
savi
ngin
terv
enti
ons
care
E:n
ephr
opat
hysc
reen
ing
(sta
ndar
dB:
AC
EI
ther
apy
anti
diab
etic
care
plus
educ
atio
n,ne
phro
path
ysc
reen
ing,
AC
EI
trea
tmen
t,re
tino
path
ysc
reen
ing)
B:re
tino
path
ysc
reen
ing
Ver
yco
st-e
ffec
tive
Inte
nsiv
elif
esty
lem
odifi
cati
onSt
anda
rdlif
esty
lere
com
men
dati
onor
noIG
T8
B:m
edic
alnu
trit
iona
lth
erap
y$1
,500
/QA
LYC
ost
savi
ng-$
84,7
00/Q
ALY
‡
inte
rven
tion
A:p
hysi
cala
ctiv
ity
Uni
vers
alop
port
unis
tic
scre
enin
gfo
run
diag
nose
dty
pe2
diab
etes
inA
fric
anA
mer
ican
sbe
twee
n45
and
54ye
ars
old
No
scre
enin
gA
fric
anA
mer
ican
sag
ed45
–54
year
s1
B$1
9,60
0/Q
ALY
$19,
600/
QA
LY
Inte
nsiv
egl
ycem
icco
ntro
las
inU
KPD
Sse
ttin
gC
onve
ntio
nalg
lyce
mic
cont
rol
Typ
e2
new
lydi
agno
sed
6A
,B
$3,4
00/Q
ALY
Cos
tsa
ving
-$12
,400
/QA
LYSt
atin
ther
apy
No
stat
inth
erap
yT
ype
2,w
ith
hype
rlip
idem
ia,
wit
hC
VD
hist
ory
3A
$2,8
00/L
YGC
ost
savi
ng-$
12,3
00/L
YG
Smok
ing
cess
atio
nN
osm
okin
gce
ssat
ion
Typ
e2
1A
,B�
$25,
000/
QA
LY�
$25,
000/
QA
LY-$
89,8
00/Q
ALY
(age
d85
–94
year
s)A
nnua
lscr
eeni
ngfo
rdi
abet
icre
tino
path
yN
osc
reen
ing
Typ
e1
2B
$2,1
50/Q
ALY
Cos
tsa
ving
-$4,
300/
QA
LY
Cost-effectiveness of diabetes interventions
1886 DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 care.diabetesjournals.org
Ann
uals
cree
ning
for
diab
etic
reti
nopa
thy
No
scre
enin
gT
ype
23
B$6
,900
/QA
LYC
ost
savi
ng-$
39,5
00/Q
ALY
Imm
edia
tevi
trec
tom
yto
trea
tdi
abet
icre
tino
path
y
Def
erra
lofv
itre
ctom
yM
ixed
popu
lati
onof
type
1an
dty
pe2
1M
enti
oned
but
not
expl
icit
lypr
ovid
edle
vel,
supp
orte
dby
tria
ls
$2,9
00/Q
ALY
$2,9
00/Q
ALY
Cos
t-ef
fect
ive
Tar
gete
dsc
reen
ing
for
undi
agno
sed
type
2di
abet
es
No
scre
enin
gU
.S.p
opul
atio
nw
ith
hype
rten
sion
45ye
ars
and
olde
r
1B:
inad
ults
ofan
yag
ew
hoar
eov
erw
eigh
tor
obes
ean
dw
hoha
veon
eor
mor
ead
diti
onal
risk
fact
ors
for
diab
etes
.In
thos
ew
itho
utth
ese
risk
fact
ors,
test
ing
shou
ldbe
gin
atag
e45
$49,
200/
QA
LY$4
6,80
0–$7
0,50
0/Q
ALY
star
ting
atdi
ffer
ent
age
Inte
nsiv
ein
sulin
trea
tmen
tC
onve
ntio
nalg
lyce
mic
cont
rol
Typ
e1
4A
,B$2
8,90
0/Q
ALY
$10,
200–
$50,
800/
QA
LY
Inte
nsiv
egl
ycem
icco
ntro
las
inth
eU
.S.s
etti
ngC
onve
ntio
nalg
lyce
mic
cont
rol
Typ
e2
new
lydi
agno
sed
at25
–54
year
sol
d
1A
,B$2
7,50
0/Q
ALY
$14,
400-
$56,
000/
QA
LY
Inte
nsiv
egl
ycem
icco
ntro
lth
roug
hlif
esty
lem
odifi
cati
on
Con
vent
iona
lgly
cem
icco
ntro
lT
ype
2ne
wly
diag
nose
d1
A,B
$33,
100/
QA
LY$3
3,10
0/Q
ALY
Stat
inth
erap
yN
ost
atin
ther
apy
Typ
e2,
wit
hhy
perl
ipid
emia
,w
itho
utC
VD
hist
ory
3A
:sta
tin
ther
apy
for
diab
etic
pati
ents
wit
hout
CV
Dw
hoar
eol
der
than
40ye
ars
and
have
one
orm
ore
othe
rC
VD
risk
fact
ors
$38,
200/
LYG
§$6
,100
/LYG
–$61
,300
/LYG
$77,
800/
QA
LY
Mul
ti-c
ompo
nent
inte
rven
tion
s(i
nten
sive
Inte
nsiv
ein
sulin
cont
rol
Typ
e1
1A
,B:i
nten
sive
insu
linco
ntro
l$4
9,80
0/LY
G(n
onU
.S.)
$46,
500-
$50,
600/
LYG
insu
linco
ntro
l,A
CE
IA
:AC
EI
ther
apy
trea
tmen
t,ey
esc
reen
ing
and
ensu
ing
trea
tmen
t)
B:ey
esc
reen
ing
Mar
gin
ally
cost
-eff
ecti
veIn
tens
ive
glyc
emic
cont
rol
asin
the
U.S
.set
ting
Con
vent
iona
lgly
cem
icco
ntro
lT
ype
2ne
wly
diag
nose
dA
llag
egr
oup
diag
nose
dof
diab
etes
at25
year
san
dol
der
1A
,B$6
2,00
0/Q
ALY
$14,
400–
$3m
illio
n/Q
ALY
Eye
scre
enin
gev
ery
2ye
ars
Eye
scre
enin
gev
ery
3ye
ars
Typ
e2
1B:
annu
aley
esc
reen
ing
reco
mm
ende
d,le
ssfr
eque
ntex
ams
(eve
ry2–
3ye
ars)
may
beco
nsid
ered
follo
win
gon
eor
mor
eno
rmal
eye
exam
s
$54,
000/
QA
LY$5
4,00
0/Q
ALY
Li and Associates
care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 1887
Not
cost
-eff
ecti
veU
nive
rsal
oppo
rtun
isti
csc
reen
ing
for
undi
agno
sed
type
2di
abet
es
Tar
gete
dsc
reen
ing
inpe
rson
sw
ith
hype
rten
sion
U.S
.pop
ulat
ion
45ye
ars
and
olde
r1
B�
$100
,000
/QA
LY$7
0,10
0-$9
28,0
00/Q
ALY
Uni
vers
alop
port
unis
tic
scre
enin
gfo
run
diag
nose
dty
pe2
diab
etes
and
ensu
ing
trea
tmen
t
No
scre
enin
gU
.S.p
opul
atio
n45
year
san
dol
der
2B
�$1
00,0
00/Q
ALY
$70,
100-
$1m
illio
n
Inte
nsiv
egl
ycem
icco
ntro
las
inth
eU
.S.s
etti
ngC
onve
ntio
nalg
lyce
mic
cont
rol
Typ
e2
New
lydi
agno
sed
at55
–94
year
s
1A
,B�
$100
,000
/QA
LY�
$100
,000
–$3
mill
ion/
QA
LY
Eye
scre
enin
gev
ery
year
Eye
scre
enin
gev
ery
2ye
ars
Typ
e2
1B
$116
,800
/QA
LY$1
16,8
00/Q
ALY
Sup
por
tive
evid
ence
Cos
tsa
vin
gSc
reen
ing
for
GD
Mw
ith
sequ
enti
alm
etho
d�N
osc
reen
ing
30-y
ear-
old
preg
nant
wom
enbe
twee
n24
–28
wee
ks’
gest
atio
n
1C
Cos
tsa
ving
Cos
tsa
ving
Scre
enin
gfo
rG
DM
wit
h10
0-g
GT
TN
osc
reen
ing
30-y
ear-
old
preg
nant
wom
enbe
twee
n24
–28
wee
ks’
gest
atio
n
1C
Cos
tsa
ving
Cos
tsa
ving
Scre
enin
gfo
rG
DM
wit
hse
quen
tial
met
hod
75-g
GT
T30
-yea
r-ol
dpr
egna
ntw
omen
betw
een
24–2
8w
eeks
’ge
stat
ion
1C
Cos
tsa
ving
Cos
tsa
ving
Scre
enin
gfo
rG
DM
wit
h10
0-g
GT
T75
-gG
TT
30-y
ear-
old
preg
nant
wom
enbe
twee
n24
–28
wee
ks’
gest
atio
n
1C
Cos
tsa
ving
Cos
tsa
ving
Dia
bete
sse
lf-m
anag
emen
ted
ucat
ion
No
educ
atio
nT
ype
11
BC
ost
savi
ngC
ost
savi
ng
Rei
mbu
rsem
ent
for
AC
EI
bypu
blic
insu
ranc
ePa
ying
out-
of-p
ocke
tT
ype
11
EC
ost
savi
ngC
ost
savi
ng
Rei
mbu
rsem
ent
for
AC
EI
bypu
blic
insu
ranc
ePa
ying
out-
of-p
ocke
tT
ype
21
EC
ost
savi
ngC
ost
savi
ng
Scre
enin
gus
ing
mob
ileca
mer
aan
del
ectr
onic
ally
tran
smit
ted
toa
data
read
ing
cent
eran
dre
adby
trai
ned
pers
onne
l
Ret
ina-
spec
ialis
tsvi
sit
Mix
edpo
pula
tion
ofty
pe1
and
type
2at
are
mot
ear
ea
1R
ecom
men
ded
but
not
leve
led,
assu
me
leve
lEC
ost
savi
ngC
ost
savi
ng
Cost-effectiveness of diabetes interventions
1888 DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 care.diabetesjournals.org
Scre
enin
gfo
rdi
abet
icne
phro
path
yan
den
suin
gA
CE
Ior
AR
Bth
erap
y
Tre
atun
til
mac
roal
bum
inur
iaT
ype
13
E:s
cree
ning
A:A
CE
Itr
eatm
ent
Cos
tsa
ving
Cos
tsa
ving
-$58
,400
/QA
LY
Inte
nsifi
edfo
otul
cer
trea
tmen
tSt
anda
rdtr
eatm
ent
Am
ixed
popu
lati
onof
type
1an
dty
pe2
1B
Cos
tsa
ving
Cos
tsa
ving
Ver
yco
st-e
ffec
tive
Inte
nsiv
edi
etan
ded
ucat
ion
Stan
dard
anti
diab
etic
care
Wom
enw
ith
GD
Mhi
stor
y,cu
rren
tly
IGT
1A
,B$2
,500
/LYG
$2,5
00/L
YG
Uni
vers
alop
port
unis
tic
scre
enin
gfo
rty
pe2
diab
etes
inyo
unge
ran
dce
rtai
net
hnic
grou
ps
No
scre
enin
gA
fric
anA
mer
ican
s,ag
ed25
–44
year
s1
B:if
over
wei
ght
orob
ese
$3,1
00/Q
ALY
$1,3
00–$
19,6
00/Q
ALY
Scre
enin
gfo
rG
DM
100-
gG
TT
Sequ
enti
alm
etho
d30
-yea
r-ol
dpr
egna
ntw
omen
betw
een
24–2
8w
eeks
’ge
stat
ion
1E
$35,
200/
QA
LYfo
rm
ater
nal
outc
omes
,$9,
000/
QA
LYfo
rne
onat
alou
tcom
es
$9,0
00–$
35,2
00/Q
ALY
Dia
bete
sse
lf-m
anag
emen
ted
ucat
ion
No
educ
atio
nT
ype
21
B$4
,000
/LYG
$4,0
00/L
YG
Dis
ease
man
agem
ent
No
dise
ase
man
agem
ent
prog
ram
Typ
e2
orm
ixed
type
s2
Men
tion
edbu
tno
tpr
ovid
edle
vel,
assu
me
leve
lE
$23,
350/
QA
LY$4
,800
–$68
,400
/QA
LYfo
rgr
oups
wit
hdi
ffer
ent
insu
ranc
e
SMBG
3ti
mes
/day
¶N
oSM
BGT
ype
2tr
eate
dw
ith
oral
agen
tsin
ala
rge
HM
O
1E
$6,9
00/Q
ALY
$540
–$30
,300
/QA
LYfo
rdi
ffer
ent
tim
eho
rizo
n
SMBG
1ti
me/
day¶
No
SMBG
1E
$9,7
00/Q
ALY
$8,2
00–$
24,2
00/Q
ALY
for
diff
eren
tti
me
hori
zon
Cos
t-ef
fect
ive
Met
form
inPl
aceb
oIG
T6
E$2
6,60
0/Q
ALY
Cos
tsa
ving
-$47
,900
/QA
LY‡
Mar
gin
ally
cost
-eff
ecti
veN
AN
otco
st-e
ffec
tive
NA
Un
cert
ain
Opt
imal
scre
enin
gfo
rty
pe2
diab
etes
star
ting
age
U.S
.pop
ulat
ion
45ye
ars
and
olde
r2
B:re
com
men
dst
arti
ngsc
reen
ing
for
type
2di
abet
esat
age
45ye
ars
ifno
othe
rri
skfa
ctor
s
AC
EI,
angi
oten
sin
conv
erti
ngen
zym
ein
hibi
tors
;A
RB
,an
giot
ensi
nre
cept
orbl
ocke
r;C
VD
,ca
rdio
vasc
ular
dise
ase;
ESR
D,
end
stag
ere
nal
dise
ase;
GD
M,
gest
atio
nal
diab
etes
;G
TT
,gl
ucos
eto
lera
nce
test
;IG
T,
impa
ired
gluc
ose
tole
ranc
e;L
YG
,life
year
gain
ed;N
A,n
otav
aila
ble;
QA
LY
,qua
lity
adju
sted
life
year
s;SM
BG
,sel
f-m
onit
orin
gbl
ood
gluc
ose.
A,a
sde
fined
inSt
anda
rds
ofM
edic
alC
are
inD
iabe
tes—
2008
:cle
arev
iden
cefr
omw
ell-
cond
ucte
d,ge
nera
lizab
le,r
ando
miz
edco
ntro
lled
tria
lsth
atar
ead
equa
tely
pow
ered
;com
pelli
ngno
n-ex
peri
men
tale
vide
nce,
i.e.,
“all
orno
ne”
rule
deve
lope
dby
the
Cen
tre
for
Evi
denc
e-Ba
sed
Med
icin
eat
Oxf
ord;
supp
orti
veev
iden
cefr
omw
ell-
cond
ucte
dra
ndom
ized
cont
rolle
dtr
ials
that
are
adeq
uate
lypo
wer
ed.
B,
asde
fined
inSt
anda
rds
ofM
edic
alC
are
inD
iabe
tes–
2008
:su
ppor
tive
evid
ence
from
wel
l-co
nduc
ted
coho
rtst
udie
s;su
ppor
tive
evid
ence
from
aw
ell-
cond
ucte
dca
se-c
ontr
olst
udy.
C,a
sdefi
ned
inSt
anda
rdso
fMed
ical
Car
ein
Dia
bete
s–20
08:s
uppo
rtiv
eev
iden
cefr
ompo
orly
cont
rolle
dor
unco
ntro
lled
stud
ies;
confl
icti
ngev
iden
cew
ith
the
wei
ghto
fevi
denc
esu
ppor
ting
the
reco
mm
enda
tion
.E,a
sdefi
ned
inSt
anda
rdso
fMed
ical
Car
ein
Dia
bete
s–20
08:e
xper
tcon
sens
usor
clin
ical
expe
rien
ce.*
,the
sam
ein
terv
enti
ons
appl
ied
todi
ffer
entp
opul
atio
nsor
com
pare
dw
ith
diff
eren
tcom
pari
son
inte
rven
tion
sw
ere
trea
ted
asdi
ffer
ents
peci
ficin
terv
enti
ons.
†,in
clud
ing
foot
exam
s,ap
prop
riat
efo
otw
ear,
trea
tmen
t,an
ded
ucat
ion.
‡,th
est
udy
for
wit
hin
tria
land
the
resu
lts
from
heal
thpl
anpe
rspe
ctiv
ear
eno
tuse
dfo
rde
term
inin
gth
eco
st-e
ffec
tive
ness
ofth
ein
terv
enti
on.§
,get
this
num
ber
byta
king
the
med
ian
for
wom
enin
one
stud
y(c
onse
rvat
ive,
wom
en�
men
)as
the
resu
lts
for
that
stud
y,th
enta
keth
em
edia
nof
the
thre
est
udy.
�,50
-gG
TT
�10
0-g
GT
T.¶
,the
evid
ence
was
very
wea
k:th
ere
was
anov
er40
%pr
obab
ility
that
the
inte
rven
tion
wou
ldco
stm
ore
than
$50,
000/
QA
LYin
alo
ng-t
erm
.
Li and Associates
care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 1889
intervention compared with usual care; 2)universal opportunistic screening for type2 diabetes in African Americans aged25–44 years compared with no screen-ing; 3) 100-g GTT compared with the se-quential screening method for detectingGDM in 30-year-old pregnant womenbetween 24–28 weeks’ gestation; 4) dia-betes self-management education for per-sons with type 2 diabetes compared withno education; 5) disease managementprograms using specialist nurse–led clin-ics to treat and control hypertension orhyperlipidemia in patients with type 2 di-abetes in a city in England or a culturallysensitive case–management training pro-gram to control diabetes and its risk fac-tors in a Latino population with both type1 and type 2 diabetes in a U.S. countycompared with usual care only; 6) self-monitoring of blood glucose (SMBG)three times per day compared with noSMBG in type 2 noninsulin users; and 7)SMBG once per day compared with noSMBG in type 2 noninsulin users. One ofthe 18 specific interventions was cost-effective, i.e., the use of metformin to pre-vent type 2 diabetes in obese persons withimpaired glucose tolerance comparedwith standard lifestyle intervention. Nointerventions in the “supportive” evi-dence category were “marginally cost-effective” or “not cost-effective.”
Current evidence is uncertain on howthe CE of screening for undiagnosed type2 diabetes would change with the age ofthose screened. Two studies evaluated theCE of screening for undiagnosed type 2diabetes; one study reported that cost-effectiveness ratios (CERs) increased withinitial screening age (16) while the otherreported that they decreased with screen-ing age (35).
CONCLUSIONS — Our systematicreview showed that, with few exceptions,ADA-recommended interventions forpreventing or treating diabetes and itscomplications were cost saving, very cost-effective, or cost-effective (i.e., with anICER of less than $50,000 per QALY orLYG), although the strength of evidencevaried. Generally, interventions that costless than $50,000 per QALY are consid-ered an efficient use of resources andworth recommending (11). Interventionswith strong evidence for being cost sav-ing, very cost-effective, or cost-effectiveshould be considered for implementa-tion. Interventions with supportive evi-dence for being cost saving, very cost-effective, or cost-effective should be
adopted if extra resources are available orif similar interventions with strong evi-dence are unavailable or infeasible in aspecific setting.
The one intervention recommendedby the ADA that was shown as not CE wasscreening for type 2 diabetes of all U.S.residents aged 45 years and older. Whenconsidering allocating resources effi-ciently, universal screening for undiag-nosed diabetes should be undertakenwith great caution. The high CE ratio foruniversal screening for undiagnosed type2 diabetes was primarily attributable tothe small gain in health benefit. For exam-ple, screening everyone aged 45 years andolder gained only 0.003 QALY per eligibleperson compared with no screening.However the additional costs associatedwith screening and early treatments wererelatively large ($564 per person). Al-though detecting and treating diabetesearlier can prevent future diabetes-relatedcomplications and their associated medi-cal costs, such savings are relatively small($57 per person). Combining the healthbenefit and costs would yield an ICER ofmore than $1 million per QALY (35). Analternative to broad screening is to focuson screening persons with additional riskfactors, such as hypertension. Such tar-geted screening is shown to be cost-effective when compared with noscreening or universal screening.
Intensive glycemic control for all U.S.residents with type 2 diabetes diagnosedat age 25 years and older is marginally CE.However the cost-effectiveness of this in-tervention varies by age at the time of thediabetes diagnosis. The intervention iscost-effective in persons diagnosed at25–54 years of age. However, intensiveglycemic control for those diagnosed withdiabetes at 55 years of age and older is notcost-effective. In fact, this result is consis-tent with the ADA’s recommendation ofless stringent A1C goals for patients withlimited life expectancies.
The ADA recommended annual eyescreening for diabetic retinopathy. Thisrecommended intervention is very cost-effective compared with no screening inpersons with type 2 diabetes. If consider-ing the efficient allocation of resources,however, screening every other yearmight be a better alternative. Screeningannually leads to a small health benefitbut results in a moderate additional cost.For example, Vijan et al. (69) showedthat, compared with a 2-year screening,annual screening among persons at mod-erate risk (65 years old with A1C level
9%) resulted in an increase of 2–3 days ofsight at a cost of $540–690 per person.However the ADA also stated in its recom-mendation that “less frequent exams (ev-ery 2–3 years) may be consideredfollowing one or more normal eyeexams.”
For the interventions with uncertainCE (including optimal age of startingscreening for type 2 diabetes), followingthe current treatment guidelines may bethe best option until more evidence ontheir CE is available.
The CEs of 43 ADA-recommendedinterventions were evaluated. Of these,25 were in the “strong” evidence category.This number would probably have beenlarger if we had used less stringent criteriato define evidence as being strong. Forexample, evidence on the CE of usingmetformin to prevent type 2 diabetesamong high-risk individuals was consid-ered “supportive” in our current classifi-cation even though the efficacy of theintervention was shown by well-conducted multi-center large clinical tri-als in different country settings (71,72),and its CE was evaluated by “excellent”CE studies (25,34). This intervention wasconsidered to have supportive evidencebecause it ranked lower in the ADA rec-ommendations (7).
Among all the interventions consid-ered, evidence for the CE of primary pre-vention through intensive lifestylemodification was the strongest regardingthe quantity and quality of the CE studiesand efficacy data. Several well-conductedclinical trials have shown the efficacy ofintensive lifestyle modification in pre-venting diabetes in different country set-tings, such as the U.S. DPP (71), FinnishDiabetes Prevention Study (73), China DaQing Diabetes Prevention Study (74), andIndian DPP (72). Eight cost-effectivenessstudies (seven of them rated as excellentquality) have been conducted by differentgroups in different countries based on datafrom these well-conducted clinical trials(15,25,34,36,41,50,59,66). The resultsfrom these studies consistently showed thatintensive lifestyle modification in personswith impaired glucose tolerance was costsaving or very cost-effective in the long run(15,25,34,36,41,50,59). Even in a short-term and one-on-one consulting setting,the intervention remained cost-effective(66). The intervention would be morecost-effective than existing studies show ifthe cost of the lifestyle intervention couldbe reduced. This might be achieved bychanging the setting in which the inter-
Cost-effectiveness of diabetes interventions
1890 DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 care.diabetesjournals.org
vention is provided. Only one studyfound a DPP-like intervention to be mar-ginally cost-effective (25). Even in thisstudy, however, the intervention wouldhave been very cost-effective (23) if donein the type of group environment that ismost likely in a real-world setting. Agroup-based, DPP-style lifestyle interven-tion partnership with the YMCA costs$275 to $325 per participant in the firstyear compared with $1,400 in the one-on-one setting of the DPP trial (75). Pre-venting diabetes, in particular by lifestylemodification, is not only effective but alsoa very efficient use of health careresources.
The CE of an intervention can vary bycountry setting. For example, intensiveglycemic control (with a goal A1C level of7%) in type 2 diabetic patients diagnosedat 25 years of age and older was margin-ally cost-effective in the U.S. but verycost-effective in other developed coun-tries. Although the efficacy data of allstudies of intensive glycemic control intype 2 diabetic patients were based on thesame UKPDS data, the cost data werebased on how residents of the differentcountries used health services and thecost of those services. The incrementalcost of intensive glycemic control wasmuch higher in the U.S. than in the U.K.because of different practice patterns. Pa-tients outside the U.S. did not receive di-abetes disease management services andhad less frequent self-testing and physi-cian office visits than their U.S. counter-parts at the time these studies wereconducted. If using the health services asdescribed in the UKPDS setting but withthe U.S. cost of these services, the CE ofthe intensive glycemic control in the U.S.would resemble that of other developedcountries.
Future economic evaluation of diabe-tes interventions should consider the fol-lowing. First, more studies are needed toevaluate the CE of interventions that fellin the “supportive” evidence category. Forstudies with weaker efficacy data, furtherefficacy studies are needed. Second, thereare also 38 interventions recommendedby the ADA but they have not been eval-uated for their CE or the studies did notmeet the inclusion criteria for our review(list is available upon request from theauthors). The CE of these interventionsshould be assessed. Third, more CE stud-ies are needed that address interventionsin real-world settings. For example, fewstudies considered attrition rate, non-compliance, and dropout rates in evaluat-
ing CE. Fourth, more studies are neededto evaluate the CE of public policychanges. Only two studies evaluated pub-lic insurance reimbursement of ACEItherapy and both found this interventionto be cost saving. Finally, the CE of mul-tiple interventions needs to be evaluated.In most real-world settings, patients re-ceive multiple interventions simulta-neously. Nearly all previous studies onlyevaluated the CE of a single intervention.
This review’s conclusions should beused with caution. First, our conclusionsare based on available information up toMay 2008. More studies have been pub-lished since then. In addition, data onboth the effectiveness and cost of an in-tervention could have changed since thetime the original study was conducted.Using the newly available data couldchange our current conclusion. For ex-ample, in our review, we concluded thatthe CE of optimal age to start screening fortype 2 diabetes was uncertain. A recentlypublished CE study on age at initiation ofscreening for type 2 diabetes, released af-ter our analysis was complete, mightchange that conclusion (76). Another ex-ample is the large decrease in costs formetformin, statins, and ACEIs. Studiesthat evaluate CE using current costs mightlook more favorably on interventions thatinclude statins and ACEIs than those re-ported here. Reevaluating the costs andbenefits of these interventions, using cur-rent-day costs, is beyond the scope of thisstudy. Second, when using the results andconclusions of our review, readers need tobe certain that terms are understood cor-rectly. For example, “intensive insulintreatment” in our review meant “multipleinsulin injection” or “insulin infusion.”Developments in medical technologymight make continuous glucose monitor-ing systems, which record blood glucoselevels throughout the day and night, morecommon. Drugs such as TZD Byetta andGliptin, not available at the time coveredby this review, are increasingly used toachieve intensive glycemic control. TheCE of treatment with these and other newdevices and drugs are unknown. New CEanalyses are needed for these new inter-ventions. Third, not everyone will neces-sarily agree with our classification criteria.Different classification criteria might havechanged some conclusions. Fourth, mostof the CE studies are based on simulationmodeling. Although good-quality simula-tion modeling can provide information ata much lower cost than clinical trials,models are based on assumptions and
represent a simplification of—and there-fore might depart from—reality. Fifth,these CE studies use different methods,which could account for some differencesin CERs. If the results from different mod-els were consistent, we would have moreconfidence in the conclusion on the CE ofthe intervention. Sixth, we used the samethreshold for the classification of the CEof interventions regardless of whether theICERs were expressed as dollars per LYGor dollars per QALY, although they aredifferent measures. The studies that re-ported costs per LYG did not incorporatethe impact of the intervention on qualityof life into the analysis. If they did, thecost per QALY could be higher, lower, orthe same depending on the relative mag-nitude of the health benefit of the inter-vention on quality of life. Seventh, theinterpretation of the CE of an interventionmust include consideration of variablessuch as study population, comparison in-terventions, and country setting. Lastly,our recommendations are based on theCE of the interventions and not their effi-cacy; therefore, these recommendations arenot necessarily the same as the ADA recom-mendations.
The importance of CE in decisionmaking should not be overstated. CE isonly one aspect to consider. CE analysisdoes not address the distribution of costsand the benefits of an intervention, soci-etal or personal willingness to pay, socialand legal aspects, or ethical issues associ-ated with each intervention. All these as-pects are important in formulating publicpolicy. The good news is that our studyshows that a majority of the recom-mended diabetes interventions provideboth health benefits and good use ofhealth care resources.
Acknowledgments— The authors con-ducted this project as part of their jobs as em-ployees of the Centers for Disease Control andPrevention (CDC). The CDC is a federalagency in the U.S. government. The authorshave no financial interest in this project.
Parts of this study were presented at the69th Scientific Sessions of the American Dia-betes Association, New Orleans, Louisiana,5–9 June 2009, and at the Division of DiabetesTranslation 2007 Annual Conference, Atlanta,Georgia, April 30–May 3, 2007.
We thank Drs. Sue Kirkman, Richard Khan,William H. Herman, John Anderson, SusanBraithwaite, Dan Lorber, and Vivian Fonsecafor reviewing the earlier version of this manu-script and providing valuable comments. Wealso thank Elizabeth Lee Greene for her invalu-able editorial assistance.
Li and Associates
care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 1891
References1. American Diabetes Association. Eco-
nomic costs of diabetes in the U.S. in2007. Diabetes Care 2008;31:1–20
2. Klonoff DC, Schwartz DM. An economicanalysis of interventions for diabetes. Di-abetes Care 2000;23:390–404
3. Raikou M, McGuire A. The economics ofscreening and treatment in type 2 diabetesmellitus. Pharmacoeconomics 2003;21:543–564
4. Zhang P, Engelgau MM, Norris SL, GreggEW, Narayan KM. Application of eco-nomic analysis to Diabetes and DiabetesCare. Ann Intern Med 2004;140:972–977
5. Vijgen SM, Hoogendoorn M, Baan CA, deWit GA, Limburg W, Feenstra TL. Costeffectiveness of preventive interventionsin type 2 diabetes mellitus: a systematicliterature review. Pharmacoeconomics2006;24:425–441
7. American Diabetes Association. Stan-dards of medical care in diabetes–2009.Diabetes Care 2008;31(Suppl. 1):S12–S54
8. Drummond MF, Jefferson TO. Guidelinesfor authors and peer reviewers of eco-nomic submissions to the BMJ. The BMJEconomic Evaluation Working Party.BMJ 1996;313:275–283
9. Federal Reserve Bank. Foreign ExchangeRates (annual) [Internet], 2008. Availableat http://www.federalreserve.gov/releases/g5a/. Accessed 30 September 2008
10. U.S. Department of Labor. Bureau of La-bor Statistics: Consumer Price Index:achieved Consumer Price Index detailedreport information: U.S. city average, bymedical care [Internet], 2008. Available athttp://www.bls.gov/cpi/cpi_dr.htm#2007.Accessed 30 September 2008
11. Laupacis A, Deeny D, Detsky AS, TugwellPX. How attractive does a new technologyhave to be to warrant adoption and utili-zation? Tentative guidelines for usingclinical and economic evaluations. CanMed Assoc J 1992;146:473–481
12. Grosse SD. Assessing cost-effectiveness inhealth care: history of the $50,000 perQALY threshold. Value Health 2008;8:165–178
13. Almbrand B, Johannesson M, Sjostrand B,Malmberg K, Ryden L. Cost-effectivenessof intense insulin treatment after acutemyocardial infarction in patients with di-abetes mellitus; results from the DIGAMIstudy. Eur Heart J 2000;217:33–39
14. Borch-Johnsen K, Wenzel H, Viberti GC,Mogensen CE. Is screening and interven-tion for microalbuminuria worthwhile inpatients with insulin dependent diabetes?BMJ 1993;306:1722–1725
15. Caro JJ, Getsios D, Caro I, Klittich WS,O’Brien JA. Economic evaluation of ther-apeutic interventions to prevent type 2 di-abetes in Canada. Diabet Med 2004;21:1229–1236
16. CDC Diabetes Cost-Effectiveness StudyGroup. The cost-effectiveness of screen-ing for type 2 diabetes. JAMA 1998;280:1757–1763
18. Clarke P, Gray A, Adler A, Stevens R,Raikou M, Cull C, Stratton I, Holman R,UKPDS Group. Cost-effectiveness analy-sis of intensive blood-glucose controlwith metformin in overweight patientswith type II diabetes (UKPDS No. 51).Diabetologia 2001;44:298–304
19. Clarke PM, Gray AM, Briggs A, StevensRJ, Matthews DR, Holman RR; UnitedKingdom Prospective Diabetes Study.Cost-utility analyses of intensive bloodglucose and tight blood pressure controlin type 2 diabetes (UKPDS 72). Diabeto-logia 2005;48:868–877
20. Clarke WF, Churchill DN, Forwell L,Macdonald G, Foster S. To pay or not topay? A decision and cost-utility analysis ofangiolensin-converting-enzyme inhibitortherapy for diabetic nephropathy. CMAJCanadian Medical Association Journal2000;162:195–198
21. Coyle D, Rodby R, Soroka S, Levin A,Muirhead N, de Cotret PR, Chen R,Palmer A. Cost-effectiveness of irbesartan300 mg given early versus late in patientswith hypertension and a history of type 2diabetes and renal disease: a Canadianperspective. Clinical Therapeutics 2007;29:1508–1523
23. Earnshaw SR, Richter A, Sorensen SW,Hoerger TJ, Hicks KA, Engelgau M,Thompson T, Narayan KM, WilliamsonDF, Gregg E, Zhang P. Optimal allocationof resources across four interventions fortype 2 diabetes. Med Decis Making 2002;22(Suppl. 5):S80–S91
24. Eastman RC, Javitt JC, Herman WH, Das-bach EJ, Copley-Merriman C, Maier W,Dong F, Manninen D, Zbrozek AS, Kotsa-nos J, Garfield SA, Harris M. Model ofcomplications of NIDDM: II. Analysis ofthe health benefits and cost-effectivenessof treating NIDDM with the goal of nor-moglycemia. Diabetes Care 1997;20:735–744
25. Eddy DM, Schlessinger L, Kahn R. Clini-cal outcomes and cost-effectiveness ofstrategies for managing people at high riskfor diabetes. Ann Intern Med 2005;143:251–264
26. Elliott WJ, Weir DR, Black HR. Cost-ef-fectiveness of the lower treatment goal (ofJNC VI) for diabetic hypertensive pa-tients. Joint National Committee on Pre-vention, Detection, Evaluation, andTreatment of High Blood Pressure. ArchIntern Med 2000;160:1277–1283
27. Gilmer TP, Roze S, Valentine WJ, Emy-Al-brecht K, Ray JA, Cobden D, Nicklasson L,Philis-Tsimikas A, Palmer AJ. Cost-effec-tiveness of diabetes case management forlow-income populations. Health ServicesResearch 2007;42:1943–1959
28. Golan L, Birkmeyer JD, Welch HG. Thecost-effectiveness of treating all patientswith type 2 diabetes with angiotensin-converting enzyme inhibitors. Ann InternMed 1999;131:660–667
29. Gozzoli V, Palmer AJ, Brandt A, SpinasGA. Economic and clinical impact of al-ternative disease management strategiesfor secondary prevention in type 2 diabe-tes in the Swiss setting. Swiss Med Wkly2001;131:303–310
30. Gray A, Raikou M, McGuire A, Fenn P,Stevens R, Cull C, Stratton I, Adler A, Hol-man R, Turner R. Cost effectiveness of anintensive blood glucose control policy inpatients with type 2 diabetes: economicanalysis alongside randomised controlledtrial (UKPDS 41). United Kingdom Pro-spective Diabetes Study Group. BMJ2000;320:1373–1378
31. Grover SA, Coupal L, Zowall H, Dorais M.Cost-effectiveness of treating hyperlipid-emia in the presence of diabetes: whoshould be treated? Circulation 2000;102:722–727
32. Habacher W, Rakovac I, Gorzer E, HaasW, Gfrerer RJ, Wach P, Pieber TR. Amodel to analyse costs and benefit of in-tensified diabetic foot care in Austria. JEval Clin Pract 2007;13:906–912
33. Herman WH, Alexander CM, Cook JR,Boccuzzi SJ, Musliner TA, Pedersen TR,Kjekshus J, Pyorala K. Effect of simvasta-tin treatment on cardiovascular resourceutilization in impaired fasting glucose anddiabetes: findings from the ScandinavianSimvastatin Survival Study. Diabetes Care1999;22:1771–1778
34. Herman WH, Hoerger TJ, Brandle M,Hicks K, Sorensen S, Zhang P, HammanRF, Ackermann RT, Engelgau MM, RatnerRE, Diabetes Prevention Program Re-search Group. The cost-effectiveness oflifestyle modification or metformin inpreventing type 2 diabetes in adults withimpaired glucose tolerance. Ann InternMed 2005;142:323–332
35. Hoerger TJ, Harris R, Hicks KA, DonahueK, Sorensen S, Engelgau M. Screening fortype 2 diabetes mellitus: a cost-effective-
Cost-effectiveness of diabetes interventions
1892 DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 care.diabetesjournals.org
ness analysis. Ann Intern Med 2004;140:689–699
36. Hoerger TJ, Hicks KA, Sorensen SW, Her-man WH, Ratner RE, Ackermann RT,Zhang P, Engelgau MM. Cost-effective-ness of screening for pre-diabetes amongoverweight and obese U.S. adults. Diabe-tes Care 2007;30:2874–2879
37. Javitt JC, Aiello LP, Chiang Y, Ferris FL3rd, Canner JK, Greenfield S. Preventiveeye care in people with diabetes is cost-saving to the federal government: impli-cations for health-care reform. DiabetesCare 1994;17:909–917
38. Javitt JC, Aiello LP. Cost-effectiveness ofdetecting and treating diabetic retinopa-thy. Ann Intern Med 1996;124:164–169
39. Jonsson B, Cook JR, Pedersen TR. Thecost-effectiveness of lipid lowering in pa-tients with diabetes: results from the 4Strial. Diabetologia 1999;42:1293–1301
40. Kiberd BA, Jindal KK. Screening to pre-vent renal failure in insulin dependent di-abetic patients: an economic evaluation.BMJ 1995;311:1595–1599
41. Lindgren P, Lindstrom J, Tuomilehto J,Uusitupa M, Peltonen M, Jonsson B, deFaire U, Hellenius ML, DPS Study Group.Lifestyle intervention to prevent diabetesin men and women with impaired glucosetolerance is cost-effective. Int J TechnolAssess Health Care 2007;23:177–183
42. Maberley D, Walker H, Koushik A,Cruess A. Screening for diabetic retinop-athy in James Bay, Ontario: a cost-effec-tiveness analysis. CMAJ 2003;168:160–164
43. Mason JM, Freemantle N, Gibson JM,New JP, SPLINT trial. Specialist nurse-ledclinics to improve control of hypertensionand hyperlipidemia in diabetes: economicanalysis of the SPLINT trial. Diabetes Care2005;28:40–46
44. Nicholson WK, Fleisher LA, Fox HE,Powe NR. Screening for gestational diabe-tes mellitus: a decision and cost-effective-ness analysis of four screening strategies.Diabetes Care 2005;28:1482–1484
45. Ortegon MM, Redekop WK, Niessen LW.Cost-effectiveness of prevention andtreatment of the diabetic foot: a Markovanalysis. Diabetes Care 2004;27:901–907
46. Palmer AJ, Weiss C, Sendi PP, Neeser K,Brandt A, Singh G, Wenzel H, Spinas GA.The cost-effectiveness of different man-agement strategies for type I diabetes: aSwiss perspective. Diabetologia 2000;43:13–26
47. Palmer AJ, Annemans L, Roze S, LamotteM, Rodby RA, Cordonnier DJ. An eco-nomic evaluation of irbesartan in thetreatment of patients with type 2 diabetes,hypertension and nephropathy: cost-ef-fectiveness of Irbesartan in Diabetic Ne-phropathy Trial (IDNT) in the Belgianand French settings. Nephrol Dial Trans-plant 2003;18:2059–2066
48. Palmer AJ, Annemans L, Roze S, Lamotte
M, Lapuerta P, Chen R, Gabriel S, CaritaP, Rodby RA, de Zeeuw D, Parving HH.Cost-effectiveness of early irbesartantreatment versus control (standard anti-hypertensive medications excluding ACEinhibitors, other angiotensin-2 receptorantagonists, and dihydropyridine calciumchannel blockers) or late irbesartan treat-ment in patients with type 2 diabetes, hy-pertension, and renal disease. DiabetesCare 2004;27:1897–1903
49. Palmer AJ, Annemans L, Roze S, LamotteM, Rodby RA, Bilous RW. An economicevaluation of the Irbesartan in DiabeticNephropathy Trial (IDNT) in a UK set-ting. J Hum Hypertens 2004;18:733–738
50. Palmer AJ, Roze S, Valentine WJ, SpinasGA, Shaw JE, Zimmet PZ. Intensive life-style changes or metformin in patientswith impaired glucose tolerance: model-ing the long-term health economic impli-cations of the diabetes preventionprogram in Australia, France, Germany,Switzerland, and the United Kingdom.Clin Ther 2004;26:304–321
51. Palmer AJ, Annemans L, Roze S, LapuertaP, Chen R, Gabriel S, Carita P, Rodby RA,de Zeeuw D, Parving HH, De Alvaro F.Irbesartan is projected to be cost and lifesaving in a Spanish setting for treatmentof patients with type 2 diabetes, hyperten-sion, and microalbuminuria. Kidney IntSuppl 2005;93:S52–S54
52. Palmer AJ, Valentine WJ, Ray JA, Roze S,Muszbek N. Health economic implica-tions of irbesartan treatment versus stan-dard blood pressure control in patientswith type 2 diabetes, hypertension andrenal disease: a Hungarian analysis. EurJ Health Econ 2007;8:161–168
53. Palmer AJ, Valentine WJ, Ray JA. Irbesar-tan treatment of patients with type 2 dia-betes, hypertension and renal disease: aUK health economics analysis. Int J ClinPract 2007;61:1626–1633
54. Raikou M, McGuire A, Colhoun HM, Bet-teridge DJ, Durrington PN, Hitman GA,Neil HA, Livingstone SJ, Charlton-MenysV, Fuller JH, CARDS Investigators. Cost-effectiveness of primary prevention of car-diovascular disease with atorvastatin intype 2 diabetes: results from the Collabora-tive Atorvastatin Diabetes Study (CARDS).Diabetologia 2007;50:733–740
55. Rosen AB, Hamel MB, Weinstein MC, Cut-ler DM, Fendrick AM, Vijan S. Cost-effec-tiveness of full medicare coverage ofangiotensin-converting enzyme inhibitorsfor beneficiaries with diabetes. Annals of In-ternal Medicine 2005;143:89–99
56. Roze S, Valentine WJ, Zakrzewska KE,Palmer AJ. Health-economic comparisonof continuous subcutaneous insulin infu-sion with multiple daily injection for thetreatment of type 1 diabetes in the UK.Diabet Med 2005;22:1239–1245
mathongkam T, Yathavong K. Cost-effec-tiveness of using angiotensin-convertingenzyme inhibitors to slow nephropathy innormotensive patients with diabetes typeII and microalbuminuria. Nephrology2001;6:71–77
58. Scuffham P, Carr L. The cost-effectivenessof continuous subcutaneous insulin infu-sion compared with multiple daily injec-tions for the management of diabetes.Diabet Med 2003;20:586–593
59. Segal L, Dalton AC, Richardson J. Cost-effectiveness of the primary prevention ofnon-insulin dependent diabetes mellitus.Health Promot Int 1998;13:197–209
60. Sharma S, Hollands H, Brown GC, BrownMM, Shah GK, Sharma SM. The cost-ef-fectiveness of early vitrectomy for thetreatment of vitreous hemorrhage in dia-betic retinopathy. Curr Opin Ophthalmol2001;12:230–234
61. Shearer A, Bagust A, Sanderson D, HellerS, Roberts S. Cost-effectiveness of flexibleintensive insulin management to enabledietary freedom in people with type 1 di-abetes in the UK. Diabet Med 2004;21:460–467
62. Souchet T, Durand Zaleski I, HannedoucheT, Rodier M, Gaugris S, Passa P, RENAALstudy. An economic evaluation of Losar-tan therapy in type 2 diabetic patientswith nephropathy: an analysis of theRENAAL study adapted to France. Dia-bete Metab 2003;29:29–35
63. Szucs TD, Sandoz MS, Keusch GW. Thecost-effectiveness of losartan in type 2 di-abetics with nephropathy in Switzerland:an analysis of the RENAAL study. SwissMed Wkly 2004;134:440–447
64. Tennvall GR, Apelqvist J. Prevention ofdiabetes-related foot ulcers and amputa-tions: a cost-utility analysis based onMarkov model simulations. Diabetologia2001;44:2077–2087
65. The DCCT Research Group. Lifetime ben-efits and costs of intensive therapy aspracticed in the Diabetes Control andComplications Trial. JAMA 1996;276:1409–1415
66. Diabetes Prevention Program ResearchGroup. Within-trial cost-effectiveness oflifestyle intervention or metformin for theprimary prevention of type 2 diabetes. Di-abetes Care 2003;26:2518–2523
67. Tunis SL, Minshall ME. Self-monitoringof blood glucose in type 2 diabetes: cost-effectiveness in the United States. Am JManag Care 2008;14:131–140
68. UK Prospective Diabetes Study Group.Cost effectiveness analysis of improvedblood pressure control in hypertensivepatients with type 2 diabetes: UKPDS 40.BMJ 1998;317:720–726
69. Vijan S, Hofer TP, Hayward RA. Cost-util-ity analysis of screening intervals for dia-betic retinopathy in patients with type 2diabetes mellitus. JAMA 2000;283:889–896
Li and Associates
care.diabetesjournals.org DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 1893
70. Wake N, Hisashige A, Katayama T, Kish-ikawa H, Ohkubo Y, Sakai M, Araki E,Shichiri M. Cost-effectiveness of intensiveinsulin therapy for type 2 diabetes: a 10-year follow-up of the Kumamoto study.Diabetes Res Clin Pract 2000;48:201–210
71. Knowler WC, Barrett-Connor E, FowlerSE, Hamman RF, Lachin JM, Walker EA,Nathan DM, Diabetes Prevention Pro-gram Research Group. Reduction in theincidence of type 2 diabetes with lifestyleintervention or metformin. N Engl J Med2002;346:393–403
72. Ramachandran A, Snehalatha C, Mary S,Mukesh B, Bhaskar AD, Vijay V, IndianDiabetes Prevention Programme (IDPP).The Indian Diabetes Prevention Pro-
gramme shows that lifestyle modificationand metformin prevent type 2 diabetes inAsian Indian subjects with impaired glu-cose tolerance (IDPP-1). Diabetologia2006;49:289–297
73. Lindstrom J, Louheranta A, Mannelin M,Rastas M, Salminen V, Eriksson J, Uus-itupa M, Tuomilehto J, Finnish DiabetesPrevention Study Group. The Finnish Di-abetes Prevention Study (DPS): lifestyleintervention and 3-year results on dietand physical activity. Diabetes Care 2003;26:3230–3236
74. Li G, Zhang P, Wang J, Gregg EW, YangW, Gong Q, Li H, Li H, Jiang Y, An Y,Shuai Y, Zhang B, Zhang J, Thompson TJ,Gerzoff RB, Roglic G, Hu Y, Bennett PH.
The long-term effect of lifestyle interven-tions to prevent diabetes in the China DaQing Diabetes Prevention Study: a 20-year follow-up study. Lancet 2008;371:1783–1789
75. Ackermann RT, Marrero DG. Adaptingdiabetes prevention program lifestyle in-tervention for delivery in the community.Diabetes Educ 2007;33:69
76. Kahn R, Alperin P, Eddy D, Borch-Johnsen K, Buse J, Feigelman J, Gregg E,Holman RR, Kirkman MS, Stern M,Tuomilehto J, Wareham NJ. Age at ini-tiation and frequency of screening todetect type 2 diabetes: a cost-effective-ness analysis. Lancet 2010;375:1324 –1326
Cost-effectiveness of diabetes interventions
1894 DIABETES CARE, VOLUME 33, NUMBER 8, AUGUST 2010 care.diabetesjournals.org