APPENDIX
APPENDIX TABLE 1. Sources of Unit Costs Associated With Treatment and Assumptions Used to Derive Estimated Annual Costs
APPENDIX TABLE 2. Baseline Characteristics and Patient Management Strategies Included in the CORE Diabetes Model (N = 329)
APPENDIX TABLE 3. Management Costs
APPENDIX TABLE 4. Direct Costs of Complications
APPENDIX TABLE 5.Utilities and Disutilities in the Diabetes CORE Model
APPENDIX TABLE 6.Cumulative Incidence of Complications
APPENDIX TABLE 7. Sensitivity Analyses for 3-Day Sensor
APPENDIX TABLE 8. Sensitivity Analyses for 6-Day Sensor
1
APPENDIX TABLE 1. Sources of Unit Costs Associated With Treatment and Assumptions Used to Derive Estimated Annual Costs
Description / SAPT / MDIUnit Cost, $a / Sources/Assumptions / Unit Cost, $a / Sources/Assumptions
Insulin pump, unit cost / 6504.75 /
Minimed Paradigm 722 Insulin Pump (Medtronic), list price
/ — / —Insulin pump, calculated annual cost / 1092.80 / Discounted 16% from $6504.75b; assumed 5-year lifespan / — / —
Sensor, unit cost / 42.00 /
Guardian REAL-Time Glucose Sensor (MMT-7002c) in Minilink Transmitter Kit (Medtronic), $1260 for 3 10-packs, list price
/ — / —Sensor, calculated annual costg / 2791.97 / Discounted 16% from $42 per sensorb; replaced every 3 days; 65% usec / — / —
Transmitter and charger, unit cost / 649.00 / Real-Time Transmitter for Paradigm Pump in Minilink Transmitter Kit (Medtronic), list price / — / —
Transmitter and charger, calculated annual cost / 545.16 / Discounted 16% from $649b; assumed 1-year lifespan / — / —
Sensor insertion device, unit cost / 14.70 / Sen-serter (MMT-7500) (Medtronic), list price / — / —
Sensor insertion device, calculated annual cost / 6.17 / Discounted 16% from $14.70b; assumed 2-year lifespan / — / —
Pump reservoir, unit cost / 3.47 / MiniMed Reservoir (Medtronic), $34.65 for box of 10, list price / — / —
Pump reservoir, calculated annual cost / 354.37 / Discounted 16% from $3.465 per reservoirb; replaced every 3 days / — / —
Pump infusion set, unit cost / 12.18 / MiniMed Paradigm Quick-Set Infusion Set (Medtronic), $121.80 for box of 10, list price / — / —
Pump infusion set, calculated annual cost / 1245.65 / Discounted 16% from $12.18 per setb; replaced every 3 days / — / —
Glucose meter, unit cost / 89.00 / OneTouchUltraLink Meter for MiniMed Paradigm Insulin Pump models 522/722 (Medtronic), list price / 89.00 / OneTouchUltraLink Meter for MiniMed Paradigm Insulin Pump models 522/722 (Medtronic), list price
Glucose meter, calculated annual cost / 37.38 / Discounted 16% from $89.00b; assumed 2-year lifespan / 37.38 / Discounted 16% from $89.00b; assumed 2-year lifespan
Test strip, unit cost / 1.16 / One Touch Ultra, $29.10 for 25 strips, average wholesale priced / 1.164 / One Touch Ultra, $29.10 for 25 strips, average wholesale priced
Test strips, calculated annual cost / 2052.32 / Discounted 16% from $1.164 per stripb; assumed use 2099 strips per yeare / 2052.32 / Discounted 16% from $1.164 per stripb, assumed use of 2099 strips per yeare
Lancet, unit cost / 0.10 / One Touch Ultrasoft Lancet, $10.32 for 100 lancets, average wholesale priced / 0.10 / One Touch Ultrasoft Lancet, $10.32 for 100 lancets, average wholesale priced
Lancets, calculated annual cost / 181.96 / Discounted 16% from $0.1032b; assumed same usage as for test strips (ie, 2099 lancets) / 181.96 / Discounted 16% from $0.1032b; assumed same usage as for test strips (ie, 2099 lancets)
Insulin, unit cost / 0.12 / Novolog, $119.58 for 10 ml vial (100 u/ml), average wholesale priced / 0.14-0.15 / NovologFlexpen, $230.94 for 5 pens each containing 3 ml insulin (100 u/ml), average wholesale priced; LantusSoloSTAR, $215.28 for 5 pens each containing 3 ml insulin (100 u/ml), average wholesale priced
Insulin, calculated annual cost / 1708.21 / Discounted 16% from $0.11958 per unitb; 46.56 units per dayf / 2470.67 / Discounted 16% from mean $0.15396 per unit and $0.14352 per unitb; 54.14 units per dayf
Provider time, first year / 744.41 / Estimates from STAR 3 52-week follow-up period using hourly rate updated to 2010 values corresponding to total compensation published by the US Bureau of Labor Statistics / 329.33 / Estimates from STAR 3 52-week follow-up period using hourly rate updated to 2010 values corresponding to total compensation published by the US Bureau of Labor Statistics
Provider time, subsequent years / 239.08 / Based on STAR 3 data, estimated cost for time spent by all providers recorded at visit 8 for SAPT armh / 202.06 / Based on STAR 3 data, estimated cost for time spent by all providers recorded at visit 8 for MDI armh
Patient time, first year / 4549.94 / Estimated cumulative patient time spent on diabetes care in STAR 3 52-week follow-up period for SAPT arm, based on wage estimates in 2010 US dollars from National Compensation Survey, US Bureau of Labor Statistics / 3433.31 / Estimated cumulative patient time spent on diabetes care in STAR 3 52-week follow-up period for MDI arm, based on wage estimates in 2010 US dollars from National Compensation Survey, US Bureau of Labor Statistics
Patient time, subsequent years / 3138.98 / Based on mean patient time at week 50 for adults in both STAR 3 treatment armsi / 3138.98 / Based on mean patient time at week 50 for adults in both STAR 3 treatment armsi
Abbreviations: SAPT, sensor-augmented pump therapy; MDI, multiple daily injections.
aAll costs are expressed in 2010 US dollars. Costs were not assigned to ComLink, cable, and Medtronic CareLink, because they were provided free of charge.
bFrom the Pharmacy Benefit Management Institute [1].
cFrom Bergenstal et al [2].
dFrom Red Book[3].
eBased on Huang et al [4], annual test strip use was 2008 in the CGM arm and 2190 in the SMBG arm among adults 25 years or older with HbA1C ≥ 7.0 percentage points at baseline. Mean annual test strip use was 2099 strips.
fIn a meta-analysis, Pickup et al [5] reported a mean reduction of 14% in insulin dose (ie, difference in total daily insulin dose equivalent to 7.58 units per day) for the insulin infusion arm compared with the insulin injection arm. Therefore, the total daily insulin dose for the MDI arm was 7.58/0.14 = 54.14 units per day and for the insulin infusion arm was 0.86 × 54.14 = 46.56 units per day. For the MDI arm, we assumed 50/50 estimated usage of rapid acting and long-acting insulin.
gBased on Bergenstal et al, figure 2 [2], a weighted average representing the frequency of sensor use in STAR 3 was computed using the midpoints representing percentage of use and the reported sample sizes. Estimated sensor use was 65%.
hOur analysis indicated that mean provider time at Visit 8, Week 39 (after pump initiation) was significantly different between treatments. Thus, we assigned treatment-specific provider costs for the second and subsequent years.
iMean patient time at week 50 (after pump initiation) did not differ significantly between the treatment groups. Therefore, we applied the same estimated cost for patient time to both groups.
1
APPENDIX TABLE 2. Baseline Characteristics and Patient Management Strategies Included in the CORE Diabetes Model (N=329)
Characteristic / Base-Case ValueAge at baseline, mean (SD), ya / 41.23 (12.19)
Duration of diabetes, mean (SD), ya / 20.23 (11.94)
Male, proportiona / 0.568
Baseline risk factors
HbA1c, mean (SD), percentage pointsa / 8.3 (0.5)
Systolic blood pressure, mean (SD), mm Hga / 121.2 (12.78)
Total cholesterol,mean (SD), mg/dLb / 181.0 (32.0)
High-density lipoprotein cholesterol,mean (SD), mg/dLb / 55.0 (14.0)
Low-density lipoprotein cholesterol, mean (SD), mg/dLb / 109.0 (28.0)
Triglycerides, mean (SD), mg/dL / 86.0 (54.0)
Body mass index, mean (SD), kg/m2a / 27.9 (5.11)
Smoker, proportiona / 0.146
Cigarettes per day, mean (SD)a / 8.10 (7.99)
Alcohol consumption, mean (SD), oz/wka / 0.90 (1.63)
Race/ethnicity, proportiona
Asian/Pacific Islander / 0.003
Black / 0.0395
Hispanic / 0.0243
White / 0.924
Complications at baseline, proportionc
Anginac / 0.019
Atrialfibrillationd / 0.024
Left ventricular hypertrophyd / 0.03
Microalbuminuriab / 0.10
Background diabetic retinopathy / 1
Patient management, proportionc
ACE inhibitor for primary preventiona / 0.43
ACE inhibitor for secondary preventionf / 0.82
Aspirin for primary preventiona / 0.34
Aspirin for secondary preventione / 0.74
Statin for primary preventiona / 0.37
Statin for secondary preventionf / 0.63
Screening, proportiong
Foot ulcer prevention program / 0.87
Screened for eye disease / 0.74
Screened for renal diseasea / 0.55
Other, proportionh,i
Reduction in incidence of foot ulcer with prevention program / 0.70
Reduction in amputation rate with foot care / 0.70
Sensitivity eye screening / 0.80
Specificity eye screening / 0.97
Sensitivity gross proteinuria screening / 0.85
Sensitivity microalbuminuria screening / 0.750
Specificity microalbuminuria screening / 0.970
Abbreviation: ACE, angiotensin-converting enzyme.
aFrom the STAR 3 adult cohort.
bFrom Nathan et al [6].
cFrom the Diabetes Control and Complications Trial Research Group [7].
dFrom Go et al [8].
eFrom Persell and Baker [9].
fFrom LaBresh et al [10].
gFrom Miller et al [6, 11].
hProportions were set to 0 for baseline myocardial infarction, peripheral vascular disease, stroke, heart failure, gross proteinuria, end-stage renal disease, proliferative diabetic retinopathy, severe vision loss, macular edema, cataract, uninfected foot ulcer, infected foot ulcer, healed foot ulcer, history of amputation, neuropathy, depression, patients receiving intensive insulin after MI, patients treated with extra ulcer treatment, patients screened for depression with and/ without complications, and improvement in ulcer healing rate with extra ulcer treatment.
iDefault values in the CORE Diabetes Model.
APPENDIX TABLE 3. Utilities and Disutilities in the Diabetes CORE Model
Health State / Utility or DisutilityActive ulcer / 0.600a,f
Amputation event / –0.109a,b
Post-amputation / 0.68a,b
Angina / 0.682a,b
Background diabetic retinopathy / 0.814d
Background diabetic retinopathy, wrongly treated / 0.814d
Cataract / 0.794a,e
Congestive heart failure / 0.633a,b
Depression, not treated / 0.6059a
Depression, treated / 0.814a
Edema event / 0
Post-edema / 0.814a
Fear of hypoglycemic event / 0
Gross proteinuria / 0.814b*
Heal ulcer / 0.814d
Hemodialysis / 0.49a,c
Ketoacidosis event / –0.0052h
Lactic acidosis event / 0
Macular edema / 0.794a,e
Major hypoglycemic events / –0.0052a,g
Microalbuminuria / 0.814b*
Minor hypoglycemic events / –0.0033
Myocardial infarction event / –0.129a,b
Post–myocardial infarction / 0.736a,b
Neuropathy / 0.624a,e
Peripheral vascular disease / 0.570a,c
Peritoneal dialysis / 0.56a,c
Proliferative diabetic retinopathy, laser-treated / 0.794a,e
Proliferative diabetic retinopathy, no laser treatment / 0.794a,e
Renal transplant / 0.762a,c
Severe vision loss / 0.734a,b
Stroke event / –0.181a,b
Post-stroke / 0.545a,b
Type 1 diabetes mellitus, no complications / 0.814a,b
Type 2 diabetes mellitus, no complications / 0.814a,b
aSee Palmer et al. [20].
bSee Clarke et al. [21].
cSee Tengs et al. [22].
dNo state-specific health utility identified. The health utility is conservatively assumed to be equivalent to complication-free utility [20].
eSee Australian Institute of Health and Welfare [23].
fSee Carrington et al. [24].
gSee National Institute for Clinical Excellence [25].
hWe assumed same disutility for aketoacidosis event as for a major hypoglycemic event.
APPENDIX TABLE 4. Management Costs
Variable / Base-Case Value, Mean, $Statin, annual costa / 64.39
Aspirin, annual costb / 12.70
Angiotensin-converting enzyme inhibitor, annual costc / 21.71
Screening for microalbuminuria12 / 22.34
Screening for gross proteinuria12 / 32.77
Eye screeningd / 75.28
aThe average wholesale price of simvastatin 20 mg was $18.90 for 90 tablets [3]. Assuming 1 tablet per day, the annual cost of generic simvastatin, discounted by 16%, was $64.39.
bThe average wholesale price of aspirin 81 mg was $1.49 for 36 tablets [3]. Assuming 1 tablet per day, the annual cost of generic aspirin was $12.70.
cThe average wholesale price of lisinopril 20 mg was $7.95 for 100 tablets [3]. Assuming 1 tablet per day, the annual cost of generic lisinopril was $21.71.
dFor the cost of eye screening (ie, Current Procedural Terminology code 92012), we used the average nonfacility reimbursement rate reported in the 2010 Medicare Physician Fee Schedule (
APPENDIX TABLE 5. Direct Costs of Complicationsa
Complication /Base-Case Value, $b / Values Obtained From Original Sources Used for Base-Case Analysis, $
Cardiovascular complications
Angina, first year / 8973 / 6024c
Angina, ≥ 2 years / 2318 / 1556c
Congestive heart failure, first year / 12,457 / 11,257d
Congestive heart failure, ≥ 2 years / 4321 / 3905d
Myocardial infarction, first year / 45,228 / 30,364c
Myocardial infarction, ≥ 2 years / 2499 / 1678c
Peripheral vascular disease, ≥ 1 years / 3407 / 3407e
Stroke, first year / 59,892 / 40,209c
Stroke, ≥ 2 years / 19,988 / 13,419c
Stroke death within 30 days / 12,985 / 8196f
Renal complicationsc,g
Hemodialysis, ≥ 1 years / 55,145 / 37,022
Peritoneal dialysis, ≥ 1 years / 55,145 / 37,022
Renal transplant, ≥ 1 years / 55,145 / 37,022
Acute events
Major hypoglycemic event / 1479 / 1087h
Minor hypoglycemic event / 0i / —
Ketoacidosis / 15,996 / 12,234j
Lactic acidosis / 10,987 / 9929§
Edema, onset / 1134 / 761c
Edema, follow up / 112 / 75c
Eye disease
Laser treatment / 1253k / 841c
Cataract operation / 5410i / 5410.13e
Blindness, year of onset / 5490 / 3686c
Blindness, subsequent years / 5490 / 3686c
Neuropathy, foot ulcer, and amputation
Neuropathy, ≥ 1 years / 554 / 372c
Amputation / 45,213 / 30,354c
Amputation prosthesis / 1625 / 1091c
Gangrene treatmentl / 5409i / 3798m
Infected ulcerl / 2917i / 2049m
Standard uninfected ulcerl / 1104i / 776m
aOnly US-based studies are reported.
bThe costs in 2007 US dollars from Pelletier et al [13], in 2000 US dollars from O’Brien et al (2003), in 2001 US dollars from Shearer et al [19], and in 1998 US dollars from Reed et al [15] were inflated to August 2010 US dollars according to the Consumer Price Index for the Medical Care.
cFrom O’Brien et al [12].
dFrom Pelletier et al [13].
eFrom DRG Expert[14].
fFrom Reed et al [15].
gWe assigned the same costs to the 3 treatment modalities. The treatments under study in STAR 3 are not expected to differentially impact the treatment modalities for end-stage renal disease.
hFrom Bullano et al [16].
iWe assumed the costs of minor hypoglycemia events, costs after cataract surgery, healed ulcer, history of amputation, and costs after healed ulcer to be $0.
jFrom Palmer et al [17].
kLaser photocoagulation and diagnostic tests are recommended for patients diagnosed with macular edema or proliferative retinopathy [12, 18]. Thus, we applied costs for proliferative retinopathy from O’Brien et al (2003) to laser treatment.
lMean monthly costs.
mFrom Shearer et al [19].
APPENDIX TABLE 6. Cumulative Incidence of Complications
Complication / SAPT / MDIEye complications, incidence (SD), %
Proliferative diabetic retinopathy / 19.3 (1.2) / 23.6 (1.3)
Cataract / 15.8 (1.1) / 15.3 (1.1)
Macular edema / 47.9 (1.5) / 49.3 (1.6)
Severe vision loss / 31.1 (1.5) / 31.7 (1.5)
Renal complications, incidence (SD), %
Microalbuminuria / 61.2 (1.6) / 66.1 (1.6)
Gross proteinuria / 38.1 (1.5) / 44.8 (1.6)
End-stage renal disease / 12.2 (1.0) / 14.4 (1.1)
Nephropathy death / 7.6 (0.9) / 9.0 (0.9)
Cardiovascular complications, incidence (SD), %
Myocardial infarction / 25.4 (1.4) / 26.9 (1.5)
Myocardial infarction death / 17.0 (1.3) / 17.9 (1.3)
Stroke event / 32.1 (1.5) / 30.4 (1.4)
Stroke death / 12.3 (1.1) / 11.7 (1.0)
Congestive heart failure / 31.4 (1.6) / 31.0 (1.5)
Congestive heart failure death / 15.8 (1.2) / 15.1 (1.1)
Peripheral vascular disease onset / 12.9 (1.0) / 14.7 (1.1)
Angina / 21.4 (1.4) / 21.3 (1.3)
Extremities complications, incidence (SD), %
Neuropathy / 90.1 (0.9) / 92.7 (0.8)
Foot ulcer / 45.2 (1.5) / 47.2 (1.6)
Recurring foot ulcer / 58.0 (4.0) / 60.9 (4.0)
Amputation from foot ulcer / 13.1 (1.2) / 13.7 (1.3)
Amputation from recurring foot ulcer / 4.2 (0.9) / 4.5 (0.8)
Abbreviations: SAPT, sensor-augmented pump therapy; MDI, multiple daily injections.
1
APPENDIX TABLE 7. Sensitivity Analyses for 3-Day Sensor
Description / SAPT / MDI / Cost per QALY, $Discounted Costs, mean (SD), $ / Discounted QALYs, mean (SD) / Discounted Costs, mean (SD), $ / Discounted QALYs, mean (SD)
Base-case analysis / 253,493 / 10.794 / 167,170 / 10.418 / 229,675
One-way sensitivity analyses
Reduce sensor use from 65% to 45% / 239,252 / 10.794 / 167,170 / 10.418 / 191,785
Increase sensor use from 65% to 85% / 267,734 / 267,565
Use of 1 test strip per sensor change with SAPT / 218,604 / 136,846
Use of 2 test strips per day with SAPT (as recommended for calibration) / 229,345 / 165,426
Use of 2 fewer test strips per day with SAPT vs MDI (ie, 3.75 test strips per day) / 240,603 / 195,379
Reduce discount rate from 3% to 0% / 415,575 / 15.594 / 283,912 / 14.923 / 196,344
Increase discount rate from 3% to5% / 192,953 / 8.797 / 124,348 / 8.526 / 253,438
Reduce base-case SAPT treatment costs by20% / 233,481 / 10.794 / 167,170 / 10.418 / 176,429
Increase base-case SAPT treatment costs by 20% / 273,506 / 282,921
Increase pump replacement period from 5 years to 8 years / 246,700 / 211,600
Reduce base-case SAPT hypoglycemia event rate by 75% / 248,325 / 10.821 / 201,263
Increase change in HbA1c level in SAPT arm from –1.0 to –1.2 percentage points / 253,839 / 10.932 / 168,815
Reduce change in HbA1c level in SAPT arm from –1.0 to –0.8 percentage points / 253,240 / 10.671 / 340,926
“Approach 1”: Apply 0.0329 utility increment associated with less fear of hypoglycemia until SAPTpatients experience 1 or more complicationsa / 253,493 / 10.895 / 181,181
“Approach 2”: Apply 0.0329 utility increment associated with less fear of hypoglycemia throughout the remaining lifetimes of SAPTpatients / 253,493 / 11.323 / 95,416
“Approach 3”: Apply 0.0329 utility increment associated with less fear of hypoglycemiathrough first year for SAPTpatients / 253,493 / 10.827 / 211,113
Two-way sensitivity analyses
Reduce sensor use from to 45% and assume use of 1 test strip per sensor change with SAPT / 204,363 / 10.794 / 167,170 / 10.418 / 98,956
Increase sensor use to 85% and assume use of 1 test strip per sensor change with SAPT / 232,845 / 174,736
Increase sensor use to 90% and change in mean HbA1c level in SAPT arm to –1.21 percentage points / 271,728 / 10.937 / 201,363
Increase sensor use to 70% and change in mean HbA1c leve3l in SAPT arm to –0.79 percentage points / 256,733 / 10.665 / 363,232
Reduce sensor use to 50% and change in mean HbA1c level in SAPT arm to –0.64 percentage points / 242,366 / 10.566 / 510,477
Reduce sensor use to 30% and change in mean HbA1c level in SAPT arm to –0.19% percentage points / 228,062 / 10.278 / –432,711
Reduce sensor use to 10% and change in mean HbA1c level in SAPT arm to –0.43 percentage points / 214230 / 10.433 / 3,249,134
Double direct costs of complications / 334,703 / 10.794 / 252,161 / 10.418 / 219,615
Choose “table values” from CORE model as alternative source to model progression for blood pressure and lipid parameters / 252,441 / 10.954 / 164,304 / 10.550 / 218,272
Add patient time costs to base case for treatment costs and indirect costs for days off work due to complicationsb / 423,597 / 10.794 / 343,919 / 10.418 / 211,991
Abbreviations: SAPT, sensor-augmented pump therapy; MDI, multiple daily injections; QALY, quality-adjusted life-years.
aBased upon the inputs for the cohort characteristics, all patients start the simulation in the state of background diabetic retinopathy and therefore, the quality-of-life difference between treatment groups was modeled by applying a 0.0329 utility increment to background diabetic retinopathy in the SAPT arm.
bPatient time costs associated with treatment and days lost from work due to complications were incorporated in both treatment arms.
APPENDIX TABLE 8. Sensitivity Analyses for 6-Day Sensor
Description / SAPT / MDI / Cost per QALY, $Discounted Costs, mean (SD), $ / Discounted QALY, mean (SD) / Discounted Costs, mean (SD), $ / Discounted QALY, mean (SD)
Base-case analysis / 230,352 / 10.794 / 167,170 / 10.418 / 168,104
One-way sensitivity analyses
Reduce sensor use from 65% to 45% / 223,231 / 10.794 / 167,170 / 10.418 / 149,159
Increase sensor use from 65% to 85% / 237,472 / 187,049
Use of 1 test strip per sensor change with SAPT / 194,388 / 72,417
Use of 2 test strips per day with SAPT (as recommended for calibration) / 206,204 / 103,855
Use of 2 fewer test strips per day with SAPT vs MDI (ie, 3.75 test strips per day) / 217,462 / 133,808
Reduce discount rate from 3% to 0% / 380,970 / 15.594 / 283,912 / 14.923 / 144,739
Increase discount rate from 3% to5% / 174,445 / 8.797 / 124,348 / 8.526 / 185,066
Reduce base-case SAPT treatment costs by20% / 214,968 / 10.794 / 167,170 / 10.418 / 127,173
Increase base-case SAPT treatment costs by 20% / 245,736 / 209,035
Increase pump replacement period from 5 years to 8 years / 223,558 / 150,030
Reduce base-case SAPT hypoglycemia event rate by 75% / 225,168 / 10.821, / 143,832
Increase change in HbA1c level in SAPT arm from –1.0 to –1.2 percentage points / 230,520 / 10.932, / 123,394
Reduce change in HbA1c level in SAPT arm from –1.0 to –0.8 percentage points / 230,255 / 10.671 / 249,880
“Approach 1”: Apply 0.0329 utility increment associated with less fear of hypoglycemia until SAPTpatients experience 1 or more complicationsa / 230,352 / 10.895 / 132,611
“Approach 2”: Apply 0.0329 utility increment associated with less fear of hypoglycemia throughout the remaining lifetimes of SAPTpatients / 230,352 / 11.323 / 69,837
“Approach 3”: Apply 0.0329 utility increment associated with less fear of hypoglycemiathrough first year for SAPTpatients / 230,352 / 10.827 / 154,517
Two-way sensitivity analyses
Reduce sensor use from to 45% and assume use of 1 test strip per sensor change with SAPT / 187,267 / 10.794 / 167,170 / 10.418 / 53,472
Increase sensor use to 85% and assume use of 1 test strip per sensor change with SAPT / 201,508 / 91,362
Increase sensor use to 90% and change in mean HbA1c level in SAPT arm to –1.21 percentage points / 239,432 / 10.937 / 139,165
Increase sensor use to 70% and change in mean HbA1c leve3l in SAPT arm to –0.79 percentage points / 231,993 / 10.665 / 262,899
Reduce sensor use to 50% and change in mean HbA1c level in SAPT arm to –0.64 percentage points / 224,791 / 10.566 / 391,168
Reduce sensor use to 30% and change in mean HbA1c level in SAPT arm to –0.19% percentage points / 217,693 / 10.278 / –359,022
Reduce sensor use to 10% and change in mean HbA1c level in SAPT arm to –0.43 percentage points / 210,742 / 10.433 / 3,008,277
Double direct costs of complications / 311,562 / 10.794 / 252,161 / 10.418 / 158,044
Choose “table values” from CORE model as alternative source to model progression for blood pressure and lipid parameters / 229,022 / 10.954 / 164,304 / 10.550 / 160,274
Add patient time costs to base case for treatment costs and indirect costs for days off work due to complicationsb / 400,455 / 10.794 / 343,919 / 10.418 / 150,420
Abbreviations: SAPT, sensor-augmented pump therapy; MDI, multiple daily injections; QALY, quality-adjusted life-years.