Available on line at www.sciencedirect.com
`
`SCIENCE@DIRECT•
`
`ELSEVIER
`
`Diabetes Research and Clinical Practice 63 (2004) 27-35
`
`llllll&.!Blll!'ll'IE~ mll!~Willl.lCIHI
`.I.\W!D)
`WNITICAll. !PlIU\all<ClB
`
`www.clscvicr.com/locatc/diabres
`
`Preference and resource utilization in elderly patients:
`InnoLet® versus vial/syringe
`
`John Shehnet a,*, Sherwyn Schwartz b , John Cappleman c, Gregory Peterson d,
`Soren Skovlund e, Lene Lytzen e, Lars Nicklas son r, John Liang r, William Lyness r,
`InnoLet® Study Group 1
`
`a 3131 Princeton Pike, Building 2B, Suite #104, Lawrenceville, NJ 08648, USA
`b Diabetes & Glandular Disease Clinic, 5107 Medical Drive, San Antonio, TX 78229, USA
`c CEC Inc., 10,000 West Colonial Drive, Suite 187, Ocoee, FL 34761, USA
`d Diagnostic & Critical Care Medicine PC, 411 Laurel Street, Suite 3275, Des Moines, IA 50314, USA
`e Novo Nordisk A/S, Sorgenfri, Denmark
`f Novo Nordisk Inc., Princeton, NJ 08540, USA
`
`Received 20 March 2003; received in revised form 8 August 2003; accepted 20 August 2003
`
`Abstract
`
`InnoLet® is a disposable insulin injection device with a large easy-to-read dial, large push button for injection, and audible
`clicks for each unit injected. This clinical trial assessed patient preference, satisfaction, and utilization of healthcare resources
`(estimated nursing care) for InnoLet and vial/syringe. Patients with diabetes mellitus (N = 79, mean age 68.2 ± 8.6 years,
`duration of diabetes 16.5 ± 10.9 years) having visual and/or motor disabilities and having difficulty (or required caregiver assis(cid:173)
`tance) for previous injections by vial/syringe were randomized to use of either InnoLet or vial/syringe for 6 weeks, then switched
`to the alternate regimen for 6 weeks. At the end of the study, utilization of healthcare resources was assessed in terms of the
`caregiver time required to assist in preparation, storage, and disposal of each device. For vial/syringe, 60% of patients required
`assistance in drawing up the appropriate dosage in the syringe, and 36% of patients required assistance when injecting insulin.
`A major portion of the patients (53%) could independently conduct injections (without nursing/caregiver assistance) during use
`of InnoLet, versus 20% for vial/syringe. As a result, mean daily nursing costs associated with the injection regimen were US$
`114 for the InnoLet device, and US$ 196 for vial/syringe ( P < 0.001 ). A majority of patients (82%) indicated a preference for
`the InnoLet device (P < 0.001).
`© 2003 Elsevier Ireland Ltd. All rights reserved.
`
`Keyword1·: Insulin delivery systems; Healthcare costs; Quality oflife; Patient independence; Patient acceptance
`
`* Corresponding author. Tel.: + 1-609-896-8050; fax: + 1-609-896-3053.
`E-mail address: jjsdiabetes@aol.com (J. Shelmet).
`1 InnoLet® Study Group: Bruce Bode, Edward Gillie, Sam Lerman, Serigo Mather, Diane Normandin, Richard Sievers and James West.
`
`0168-8227/$ - see front matter© 2003 Elsevier Ireland Ltd. All rights reserved.
`doi: I 0.10 l 6/j.diabres.2003.08.013
`
`Sanofi Exhibit 2141.001
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`28
`
`J. She/met et al. I Diabetes Research and Clinical Practice 63 (2004) 27-35
`
`1. Introduction
`
`The Diabetes Control and Complications Trial
`demonstrated
`that
`intensive
`insulin
`therapy can
`significantly reduce the incidence of late diabetic
`complications, and delay the progression of existing
`conditions in type 1 diabetes [1]. The United Kingdom
`Prospective Diabetes Study reported similar findings,
`and concluded that intensive therapy reduces the risk
`of many complications of type 2 diabetes (retinopa(cid:173)
`thy, nephropathy, and neuropathy) [2]. Significant
`costs are accrued by patients with diabetes who have
`poor glycemic control with insulin [3]. Diabetes is
`associated with blindness and end stage renal disease.
`Coronary artery disease is more severe in patients
`with diabetes.
`Many patients with diabetes using insulin injection
`may have poor treatment compliance. Lack of dia(cid:173)
`betes education is often a major factor. Other compli(cid:173)
`ance considerations may be poverty, fear of needles,
`denial, and lifestyle. In elderly patients, diabetes is of(cid:173)
`ten complicated by debilitating co-morbidities (vision
`problems, neuropathy, etc.) that aggravate the difficul(cid:173)
`ties of self-injection and increase the risk of dosing er(cid:173)
`rors. Any treatment approach that allows the patient to
`more easily self-inject insulin may have implications
`regarding the financial burden of diabetes complica(cid:173)
`tions, deterioration of quality of life, and requirements
`for nursing assistance.
`The InnoLet® insulin injection device was engi(cid:173)
`neered for ease of use. The disposable device has a
`large, easy-to-read dial, which aids in accurate insulin
`dose selection by the patient. The large injection button
`requires less physical effort for a patient with motor
`impairment (stroke, Parkinson's disease). InnoLet has
`been previously studied in vision-impaired patients:
`patients with visual acuities of 20/100 to 20/200 were
`able to correctly dispense an insulin dose with Inno(cid:173)
`Let, while 15 and 36% were unable to perform this
`task with a Humulin pen or vial/syringe, respectively
`[ 4].
`This study was designed to determine whether the
`InnoLet injection device provides benefit to elderly
`patients in terms of nursing resources, treatment
`satisfaction, and patient preference as compared to
`the vial/syringe method of insulin injection. Con(cid:173)
`tinued quality of care (glycemic control) was also
`monitored.
`
`2. Methods
`
`2.1. Study design
`
`This trial was performed in accordance with the
`Declaration of Helsinki, and written informed consent
`was obtained from all patients. This was a multicenter,
`randomized, open-label, two-period crossover trial in
`patients with type 1 or type 2 diabetes mellitus con(cid:173)
`ducted at 11 sites in the United States. The enrolled
`patients were currently injecting their own insulin (ei(cid:173)
`ther Novolin N or Novolin 70/30), but had difficul(cid:173)
`ties (requiring some assistance of a nurse or caregiver)
`due to motor dysfunction (arthritis, familial tremor,
`Parkinson's disease, stroke-induced partial paralysis)
`and/or visual problems (partially blind, cataracts, vi(cid:173)
`sual field defects, blurred vision). Enrolled patients
`were familiar with either a pen and/or vial and sy(cid:173)
`ringe method of insulin injection. Patients were ran(cid:173)
`domly assigned use oflnnoLet with N ovoFine® 30G x
`8 mm needles or the conventional disposable 0.5 cc
`Becton-Dickinson insulin syringe with MicroFine 30G
`permanently attached needle for the first period of 6
`weeks in duration, and switched to the alternate device
`for the second period of 6 weeks. The insulin dose
`was predetermined based on the patient's prior daily
`insulin requirements, and was changed over the course
`of study based on the health needs and/or diet of the
`patient. In this study, the costs of training patients in
`the use oflnnoLet® were negligible [ 4].
`
`2.2. Healthcare resource utilization
`
`Healthcare resource utilization for vial/syringe or
`device use was examined at the end of each study
`period: a questionnaire was utilized to determine the
`aspects of use that required assistance (i.e. what per(cid:173)
`centage of patients required to draw up the appro(cid:173)
`priate volume, to inject themselves, dispose of the
`vial/syringe, etc.). For each means of injection, the
`nursing/caregiver resource utilization was estimated as
`minutes per day necessary for assisting injection ( from
`injection preparation to disposal): 0 min was indicated
`if the patient was capable of performing all the activi(cid:173)
`ties by themselves. The resource utilization (in dollar
`amounts) was also calculated in terms of the number
`of visits per day that nurses/caregivers had to make
`to the patient in order to assist with insulin injections.
`
`Sanofi Exhibit 2141.002
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`J. She/met et al. I Diabetes Research and Clinical Practice 63 (2004) 27-35
`
`29
`
`The time necessary for each component (preparation
`to disposal) was individually tabulated by treatment
`and study period. If a patient required assistance with
`any aspect of preparation or injection, they were con(cid:173)
`sidered as requiring a nursing visit. Cost calculations
`were based upon an assumption of $80 per visit (mini(cid:173)
`mum of 1 hr per visit). Comparison of the two devices
`and the differences in dollar values were made using
`general linear models for the crossover design.
`
`2.3. Preference and acceptance questionnaires
`
`Patient preference was assessed by a questionnaire
`at the end of the study: "If you were given the choice
`of using one of the two systems (vial and syringe or
`InnoLet), which would you prefer?" Ease of use was
`also assessed: "Compared with the vial and syringe
`you were using, injection of insulin with the InnoLet
`system you are now using is easier, the same or more
`difficult?" A patient handling and acceptance ques(cid:173)
`tionnaire were provided at the end of each 6-week
`treatment period.
`
`2. 4. Efficacy and safety assessments
`
`The glycemic control was assessed by measur(cid:173)
`ing the serum fructosamine levels at the time of
`screening and at the end of each treatment period.
`Safety assessments were based on adverse events
`and adverse device effects, hypoglycemic episodes,
`vital signs, physical examination and the physicians
`review of the blood glucose (BG) diaries. Patients
`recorded meter-measured BG values and symptoms
`of hypoglycemia associated with blood glucose meter
`readings.
`Hypoglycemia was defined as minor when the pa(cid:173)
`tient had a symptom of hypoglycemia (i.e., palpita(cid:173)
`tions, tiredness, sweating, strong hunger, dizziness,
`tremor, etc.) confirmed by blood glucose meter read(cid:173)
`ing <50 mg/dl, and was able to deal with the episode
`without assistance. A major or severe hypoglycemic
`episode was an event that had a blood glucose meter
`reading <50 mg/dl and required third-party assistance.
`
`val was bigger than 50%, the null hypothesis was re(cid:173)
`jected, and it was concluded that the patients preferred
`the InnoLet device. A hypothesis of equal treatment
`median was tested for two treatments, versus the hy(cid:173)
`pothesis that they were not equal. A two-sided test was
`used when analyzing the comparison. The score data
`of the questionnaires was not assumed to be a nor(cid:173)
`mal distribution; therefore a non-parametric analysis,
`Wilcoxon Signed Rank test [5] was used for analysis
`of resource utilization (in dollars), and diabetes treat(cid:173)
`ment satisfaction. A binomial test was used for patient
`preference.
`
`3. Results
`
`3.1. Demographic and other baseline
`characteristics
`
`Seventy-nine patients (age 47-85, inclusive) with
`type 1 or type 2 diabetes and requiring ::::;50 U of in(cid:173)
`sulin per injection were randomized to the treatment
`sequence InnoLet ➔ vial/syringe or vial/syringe ➔
`Innolet. A total of73 (92%) patients out of the 79 com(cid:173)
`pleted the study. Five (6%) patients discontinued In(cid:173)
`noLet use due to adverse events, non-compliance with
`protocol, and change of insulin to Lantus. One (1 % )
`patient discontinued use of syringe due to unavailabil(cid:173)
`ity for follow-up. The demography of randomized pa(cid:173)
`tients is summarized in Table 1.
`
`3.2. Insulin regimen
`
`The insulin dose was based upon prior daily insulin
`requirements, and was changed as necessary over the
`course of the study. For all patients who completed
`study treatment, the mean± S.E.M. total daily in(cid:173)
`sulin dose at screening was 44.9 ± 0.14 IU/day (range
`4-100 IU) for patients using the Irmo Let device, and
`46. 7 ± 0.12 IU/day (range 10-100 IU) for patients us(cid:173)
`ing the vial/syringe.
`
`3.3. Healthcare resource utilization
`(assistance needed)
`
`2.5. Statistical analysis
`
`For preference data, a two-sided 95% confidence
`interval was calculated. If the lower limit of the inter-
`
`The resource utilization questionnaire was com(cid:173)
`pleted at the end of each trial period. In response to the
`question "When using a vial and syringe to administer
`
`Sanofi Exhibit 2141.003
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`30
`
`J. She/met et al. I Diabetes Research and Clinical Practice 63 (2004) 27-35
`
`Table I
`Demographics and baseline characteristics
`
`Number randomized, N
`Age (year), mean ± S.D.
`
`Sex, n (%)
`Male
`Female
`
`Race, n (%)
`Caucasian
`Black
`Hispanic
`
`BM! (kg/m2
`), mean ± S.D.
`AIC (%), mean ± S.D.
`
`Years since diagnosis
`Mean± S.D.
`
`Type of diabetes, n (%)
`Type I
`Type 2
`
`Overall
`
`79
`68.2 ± 8.6
`
`33 (42)
`46 (58)
`
`65 (82)
`6 (8)
`8 (10)
`33 ± 6.5
`7.5 ± 1.4
`
`16.5 ± 10.9
`
`2 (3)
`77 (97)
`
`insulin the patient needs my assistance", 60% of the
`patients required assistance in drawing up the appro(cid:173)
`priate volume, and 36% patients required assistance
`in injecting themselves with insulin. In response to
`the question "When using a vial and syringe to ad(cid:173)
`minister insulin the patient is totally self-sufficient",
`25% of patients were self-sufficient (able to prepare,
`store, and dispose insulin without nursing assistance)
`using vial/syringe, but using various injection aids
`(magnifying glass etc.).
`
`3.4. Resource utilization-time in minutes for
`nursing/caregiver resources
`
`The time (minutes) necessary for assisting injection
`(from injection preparation to disposal) was assessed
`by treatment and study period. The mean time spent
`by nurses or caregivers while assisting in injection
`preparation was less for patients using the InnoLet
`device (4.2 ± 8.1 min) than for vial/syringe (5.8 ±
`8.9 min).
`
`3.5. Resource utilization-number of visits for
`nursing/caregiver resources
`
`ing/caregiver assistance (Table 2). Some patients in
`the study required as many as three visits per day at
`US$ 80 per visit. For vial and syringe users, a major(cid:173)
`ity of the patients required such assistance (three daily
`visits). Only 20% of the patients with vial/syringe
`reported independence (required no assistance). The
`mean daily cost was US$ 196 per day for vial/syringe.
`By comparison, 53% of the patients were independent
`of nursing/caregiver assistance for injections with In(cid:173)
`noLet use, and the mean daily cost of their injections
`was US$ 114. Overall, the mean daily nursing cost
`of the InnoLet was significantly less than vial and
`syringe (US$ 114 versus US$ 196, P < 0.001).
`
`3.6. Patient preference, handling, and acceptance of
`the insulin delivery system
`
`Overall patient preference is summarized in Fig. 1.
`Patient preference was assessed by the question: "If
`you were given the choice of using one of the two
`systems (vial and syringe and InnoLet), which would
`you prefer?" A majority of patients (82% P-value <
`0.001) indicated a preference for the InnoLet device;
`only 8% indicated no preference in response to this
`question.
`Patient handling and acceptance of the devices was
`based upon the patient handling and acceptance ques(cid:173)
`tionnaire. Patient assessment of their handling expe(cid:173)
`rience with InnoLet is summarized in Fig. 2. Patients
`were asked, "Given your disability, how would you
`rate the InnoLet compared to vial and syringe: easier
`
`lnnolet
`(82%)
`
`The resource utilization was tabulated as a per(cid:173)
`centage of patients requiring/not requiring nurs-
`
`Fig. 1. Overall patient preference: "If you were given the choice
`of using one of the two systems (vial/syringe and lnnoLet), which
`would you prefer?".
`
`Sanofi Exhibit 2141.004
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`J. She/met et al. I Diabetes Research and Clinical Practice 63 (2004) 27-35
`
`31
`
`Table 2
`Resource utilization: proportion of patients requiring nursing/caregiver assistance
`
`No. of patients requiring assistance" (%)
`
`lnnolet (N = 72)
`
`Vial/syringe (N = 76)
`
`Steps of insulin injection
`Store supplies
`Prepare individual dosing
`Deliver individual dosing
`Attach needle to device
`Eliminate air from the needle
`Monitor subject's drawing up/dialing up of correct dose
`Prepare the site for injection (alcohol swab, pinch skin, etc.)
`Insure the patient has dosed themselves properly
`Return of supplies to the refrigerator
`Dispose of needles (lnnoLet)
`Dispose of syringe
`Dispose of vial
`Dispose of lnnoLet
`
`Overall independence
`Patients requiring nursing/caregivers time
`Patients not requiring nursing/caregivers time
`
`17 (24)
`16 (22)
`13 (18)
`15 (21)
`9 (13)
`19 (26)
`11 (15)
`16 (22)
`18 (25)
`8 (11)
`NIA
`NIA
`17 (24)
`
`34 (47)
`38 (53)
`
`18 (24)
`46 (61)
`37 (49)
`NIA
`33 (43)
`23 (30)
`13 (17)
`13 (17)
`19 (25)
`NIA
`24 (32)
`18 (24)
`NIA
`
`61 (80)
`15 (20)
`
`N/A: not applicable.
`a Times required for each step were collected. Patients having an estimated caregiver time that did not equal "0" min for that step are
`tabulated.
`
`to use, the same or harder to use?" A majority of the
`patients (82%) rated InnoLet easier to use. Only 4%
`found InnoLet harder to use, whereas 12% found In(cid:173)
`noLet and the vial/syringe to be the same.
`Patient acceptance of InnoLet is summarized in
`Fig. 3. Patients were asked: "Just after changing to
`InnoLet, how did you find managing the practical
`aspects (dosing and injecting) of the new insulin
`system?" About 86% of the patients reported that
`
`100
`
`~
`
`::R
`~ 80
`2
`co 60
`Cl:'.
`OJ
`C 40
`=o
`C co 20
`0
`
`I
`
`Easier
`
`Same
`
`Harder
`
`Fig. 2. Patient handling questionnaire. At the end of each study pe(cid:173)
`riod, patients were asked to complete a patient handling question(cid:173)
`naire. "Given the disability, rate lnnoLet compared to vial/syringe."
`Possible responses: (1) easier to use, (2) the same, and (3) harder
`to use.
`
`they found "managing the practical aspects" of in(cid:173)
`sulin injection ( e.g. dosing and injecting) by InnoLet
`to be "easy" or "very easy" when initiating insulin
`treatment, whereas 1 % of the patients found the use
`of InnoLet "difficult" or "very difficult" (Fig. 3A).
`When patients judged the practical aspects of insulin
`administration after switching to InnoLet, 97% of the
`patients considered the use of the InnoLet to be "easy"
`or "very easy," while 3% of the patients found the use
`of InnoLet to be "difficult" or "a little difficult" to use
`(Fig. 3B).
`When asked about the reliability of insulin injec(cid:173)
`tion, 62% rated Innolet as being more reliable than
`vial/syringe, 34% considered the treatment devices to
`be "about the same." Only 3% of patients considered
`InnoLet to be less reliable (Fig. 3C).
`Patients were asked: "Were you able to inject your(cid:173)
`self with less nursing assistance when using InnoLet?"
`A majority (84%) of patients indicated that they re(cid:173)
`quired less assistance when using InnoLet, 3% re(cid:173)
`quired more assistance, and 8% required about the
`same nursing assistance (Fig. 3D).
`Patients showed a positive response to the question:
`"How did you find the setting of the insulin dose with
`
`Sanofi Exhibit 2141.005
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`32
`
`J. She/met et al. I Diabetes Research and Clinical Practice 63 (2004) 27-35
`
`100
`
`80
`
`Cl)
`
`c
`Q) 60
`~
`a..
`~ 40
`
`0
`
`Very
`Easy
`
`Easy
`
`Little Difficult Very
`Difficult
`Difficult
`
`(B)
`
`20
`
`0
`
`Very
`Easy
`
`Easy
`
`Little Difficult Very
`Difficult
`Difficult
`
`100
`
`100
`
`80
`
`Cl)
`
`c
`Q) 60
`~
`a..
`~ 40
`
`0
`
`20
`
`0
`
`(A)
`
`100
`
`Q)
`
`Cl) 80
`c
`~ 60
`a..
`~
`0
`40
`
`20
`
`0
`
`80
`
`Cl)
`
`Q)
`
`c
`~ 60
`a..
`~
`0
`
`40
`
`20
`
`0
`
`Less
`
`Same
`
`More
`
`(C)
`
`More Reliable Same
`
`Less Reliable
`
`(D)
`
`100
`
`80
`
`.l!l
`C
`Q)
`
`~ 60
`a..
`~
`0
`
`40
`
`20
`
`0
`
`100
`
`.l!l 80
`
`C
`Q)
`~ 60
`a..
`~ 0
`
`40
`
`20
`
`0
`
`(E)
`
`Easy
`
`Difficult
`
`Very Difficult
`
`(F)
`
`No
`
`Sometimes Not Aware
`
`Fig. 3. Patient acceptance questionnaire. Answers to questions were scaled 0-100 (representing the least and most acceptance, respectively).
`(A) Initial experience-managing practical aspects: "Just after changing to InnoLet, how did you find managing the practical aspects
`( dosing and injecting) of the new insulin system?" Possible responses: (I) very easy and I learned the procedure very quickly, (2) easy,
`(3) a little difficult but I got used to it, (4) difficult, or (5) very difficult. (B) Current experience-managing practical aspects: "How do
`you now find the practical aspects of insulin administration after switching to InnoLet?" Possible responses: (I) very easy, (2) easy, (3)
`little difficult, (4) difficult, or (5) very difficult. (C) Reliability injecting insulin: "In terms of reliability of injecting your insulin, rate the
`lnnoLet." Possible responses: (I) more reliable than vial and syringe, (2) about the same, (3) less reliable. (D) Nursing assistance: "Were
`you able to inject yourself with less nursing assistance when using Irmo Let?" Possible responses: (I) required less nursing assistance, (2)
`required about the same nursing assistance, (3) required more nursing assistance. (E) Setting insulin dose: "How do you find setting of
`the insulin dose with InnoLet?" Possible responses: (I) easy, (2) difficult but I can manage, (3) very difficult and caused some problems
`for me. (F) Drawing incorrect insulin dose: "Did you ever draw up an incorrect insulin dose?" Possible responses: (I) no, never, (2) yes,
`sometimes, (3) I am not aware of it. (G) Pain of injection: "How did you find the injections?" Possible responses: (I) no pain at all, (2)
`slightly painful, (3) painful, (4) very painful, or (5) severely painful, hardly tolerable. (H) Self-injection: "Injecting myself .... " Possible
`responses: (I) has become a part of my daily life, (2) I dislike it but I do it without problems, (3) I do it with reluctance.
`
`Sanofi Exhibit 2141.006
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`J. She/met et al. I Diabetes Research and Clinical Practice 63 (2004) 27-35
`
`33
`
`100
`
`80
`
`Cl)
`
`c
`Q) 60
`~
`a..
`~ 40
`
`0
`
`100
`
`80
`
`Cl)
`
`c
`Q) 60
`~
`a..
`~ 40
`
`0
`
`20
`
`0
`
`(G)
`
`No
`Pain
`
`Slight Painful
`Pain
`
`Very Severely
`Painful Pain
`
`(H)
`
`20
`
`0
`
`Become part
`of life
`
`Dislike but
`do it with with
`no problem
`
`Do with
`reluctance
`
`Fig. 3. (Continued).
`
`Innolet?" Setting the insulin dosage with InnoLet was
`considered to be "easy" by 96% of patients (Fig. 3E).
`Patient acceptance of InnoLet was likewise high for
`the question: "Did you ever draw up an incorrect in(cid:173)
`sulin dose?" A majority of the patients (86%) reported
`that they had "never" drawn an incorrect dose, as com(cid:173)
`pared to 10% who sometimes drew an incorrect dose
`(Fig. 3F).
`When judging the pain of injections using InnoLet,
`a majority (73%) of the patients reported no pain at all,
`as compared to 26% who found Innolet to be slightly
`painful (Fig. 3G).
`When asked to evaluate their reaction to self(cid:173)
`administering daily injections, most patients (81 %)
`said that self-administration had "become a part of
`their daily life." Twenty percent patients said they
`"dislike it, but had no problem with the delivery
`system" (Fig. 3H).
`
`3. 7. Glycemic control
`
`During treatment of enrolled patients, no change
`in the glycemic control was observed during the en(cid:173)
`tire study period, as measured by fructosamine levels.
`For all patients who completed study treatment, the
`mean± S.D. for the serum fructosamine at screening
`was 268 ± 74.8 µmol/1, and at the end of the study
`was 269 ± 69 .4 µmol/1.
`
`3.8. Safety evaluation
`
`Forty-six treated patients reported 97 treatment(cid:173)
`emergent adverse events. Two (3%) patients of78 dur-
`
`ing InnoLet use reported four serious adverse events
`( chronic obstructive airway disease, cardiac failure,
`coagulation disorder, hypoglycemic coma) and 4 (5%)
`patients of 76 during vial/syringe treatment reported
`four serious adverse events (pneumonia, heart mur(cid:173)
`mur, hematemesis/ketosis, hyponatremia). There were
`no events that were considered by the investigator to
`be related to the study treatment.
`
`3.9. Adverse device effects
`
`Two adverse device effects were reported during the
`study (malfunctioning of the InnoLet device). These
`events were not associated with any adverse events.
`Two patients reported drippage from the needle fol(cid:173)
`lowing injection, possibly related to injection tech(cid:173)
`nique. No adverse device events were reported for use
`of vial/syringe.
`
`4. Discussion
`
`The difficulties of self-injection can be an obstacle
`to initiating insulin therapy at any age. Such difficul(cid:173)
`ties may be physical as well as psychological, with
`many people with type 2 diabetes being introduced
`to insulin self-injection at an advanced age. Some pa(cid:173)
`tients are uncomfortable with injection devices that
`appear technical or complex and some conditions such
`as poor eyesight or manual difficulties may pose phys(cid:173)
`ical obstacles to patient use of most injection devices
`previously available [6]. The likelihood of arthritis or
`tremors increases with advancing age [7] , adding to
`
`Sanofi Exhibit 2141.007
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`34
`
`J. She/met et al. I Diabetes Research and Clinical Practice 63 (2004) 27-35
`
`the difficulty of manually operating handheld devices.
`Clinical studies have shown that patients using a vial
`and syringe have more risk of drawing and injecting
`an inaccurate dose of insulin. In a study of patients
`with diabetes over the age of 40 years using tradi(cid:173)
`tional vial and syringe, Kesson and Bailie [8] found
`a relative error of about 19% in the accuracy of the
`doses drawn. The inaccuracy associated with patients
`injecting insulin can be a significant clinical problem,
`especially in the elderly. Puxty et al. [9] studied a
`population of syringe users over the age of 55 years
`(mean 66 years), their results showed a mean varia(cid:173)
`tion in dose of 12% when drawing up and expelling
`a dose of 20U. Sammler [10] found that 44% of sy(cid:173)
`ringe users over the age of 65 years injected doses of
`insulin that were more than 25% different from that
`prescribed. If errors in dose selection should occur,
`these errors may have profound consequences for pa(cid:173)
`tient safety. Hence, a large, easy-to-read dial could be
`an advantage. If patients believe that assistance from
`caregivers will be necessary, a reluctance to become
`dependent upon others may be an additional barrier to
`initiating insulin therapy [6].
`Complex insulin injection procedures, possibility of
`patient errors in drawing the insulin dose prescribed
`by the physician, and compliance with the treatment
`are perhaps the key points of insulin treatment safety
`in elderly patients with diabetes. The InnoLet insulin
`delivery system was developed to enable more pa(cid:173)
`tients to safely self-administer insulin, by providing
`a device that would be as intuitive and easily manip(cid:173)
`ulated as possible. The disposable InnoLet injection
`device provides a large "oven timer" scale, a large
`easily-handled injection button. The utility of such
`a simplified injection device was best assessed in a
`subset of patients having the most difficulties with
`self-injection. The aim of the present study was there(cid:173)
`fore to compare Innolet with vial/syringe in terms of
`utilization of healthcare resources, patient preference,
`and patient acceptance in elderly patients with dia(cid:173)
`betes who had difficulty in taking insulin injections
`without assistance.
`This clinical trial indicated that for the elderly dis(cid:173)
`abled patient population of this study (where about
`60% needed assistance in drawing up appropriate vol(cid:173)
`ume, 36% needed assistance making self-injections), a
`major portion of the patients (53%) did not need assis(cid:173)
`tance if they were provided with an Innolet injection
`
`system. This increased patient self-sufficiency com(cid:173)
`pared very favorably to the 20% who could manage
`alone with vial/syringe, and resulted in a significant
`reduction in costs of caregiver assistance (from US$
`196 to US$ 114 a day). In view of the fact that 82%
`of patients preferred the InnoLet device and consid(cid:173)
`ered it easier to use, InnoLet appeared to have quality
`of life implications as well.
`Fox et al. [4] recently evaluated patients with in(cid:173)
`sulin nai:ve type 2 diabetes and visual impairment
`(20/40 to 20/200) in terms of their preference and
`ability to operate three different insulin delivery sys(cid:173)
`tems: InnoLet, Humulin® NPH Pen (Eli Lilly), or
`vial/syringe (Becton-Dickinson). These investigators
`found that 87% of the patients had an overall prefer(cid:173)
`ence for use oflnnoLet, 13% preferred Humulin Pen
`(P < 0.001), and no patients preferred the syringe.
`In the United States, poor treatment compliance by
`patients with diabetes may often be due to a lack of
`educational support and the time constraints in the of(cid:173)
`fice of the primary care provider. Use of InnoLet could
`facilitate the initiation of insulin in more patients.
`Insulin delivery systems designed to simplify accu(cid:173)
`rate, reliable insulin delivery for people with visual
`impairment can improve the ability of such patients
`to repeatedly set and self-deliver the correct insulin
`dose. This study is additional evidence that an insulin
`injection delivery device can provide greater positive
`patient feelings than the traditional vial/syringe.
`
`5. Conclusions
`
`The InnoLet insulin injection device could reduce
`the cost of treatment by fostering independence in
`patients who had difficulties in self-injection using
`vial/syringe. Patients showed a preference for use of
`the InnoLet insulin injection device over the vial and
`syringe, with a majority of the patients reporting more
`confidence that the correct amount of insulin was
`injected using the InnoLet device. Elderly patients
`showed a higher treatment acceptance during use of
`InnoLet.
`
`Acknowledgements
`
`This study was supported by Novo Nordisk Pharma(cid:173)
`ceuticals Inc., Princeton, NJ. The authors wish to thank
`
`Sanofi Exhibit 2141.008
`Mylan v. Sanofi
`IPR2018-01676
`
`

`

`J. She/met et al. I Diabetes Research and Clinical Practice 63 (2004) 27-35
`
`35
`
`the site coordinators for their assistance during the
`trial: Kelly McCulloch, Karen Jones, Carrie-Ann Sil(cid:173)
`via, Joyce Stathopoulos, Stacy Frerichs, Debbie Kee,
`Billie Jo Shelmet, Paula Wells and Cindy Nelson.
`
`References
`
`[1] The Diabetes Control and Complications Trial Group,
`The effect of intensive
`treatment of diabetes and the
`development and progression of long-term complications in
`insulin-dependent diabetes mellitus, N. Engl. J. Med. 329
`(1993) 977-986.
`[2] United Kingdom Prospective Diabetes Study Group,
`Intensive blood-glucose control with sulphonylureas or
`insulin compared with conventional treatment and risk of
`complications in patients with type 2 diabetes, Lancet. 1998,
`352:837-853.
`[3] R.J. Rubin, W.A. Altman, D.N. Mendelson, Health care
`expenditures for people with diabetes mellitus, 1992, J. Clin.
`Endocrinol. Metab. 78 (1994) 809A-809F.
`
`[4] C. Fox, C. McKinnon, A. Wall, S.A. Lawton, Ability to
`handle, and patient preference for, insulin delivery devices in
`visually impaired patients with type 2 diabetes, Pract. Diab.
`Int. 19 (4) (2002) 104-107.
`[5] W.J. Conover, Practical nonparametric statistics, in: Wiley
`Series in Probability and Statistics: Applied Probability
`and Statistics Section, third ed., Wiley, New York, 1999,
`pp. 352-369.
`[6] E. Bashoff, R. Beaser, Insulin therapy and the reluctant
`patient, Postgrad. Med. 97 (1995) 86-90.
`[7] M. Frey, E. Barrett-Connor, P. Sledge, et al., The effect of
`non-insulin dependent diabetes mellitus on the prevalence of
`clinical osteoarthritis. A population based study, J. Rheumatol.
`23 (1996) 716-722.
`[8] C.M. Kesson, G.R. Bailie, Do patients with diabetes
`inject accurate doses of insulin? Diabetes Care 4 (1981)
`333.
`[9] J.A.H. Puxty, D.H. Hunter, W.A. Burr, Accuracy of insulin
`injection in elderly patients, Br. Med. J. (Clin. Res. ed.) 287
`(1983) 1762.
`[10] A. Sammler, Probleme alterer, insulinspritzender diabetiker
`be! deer ambulanten behandlung, Versiccherungsmedizin 42
`(1990) 59-64.
`
`Sanofi Exhibit 2141.009
`Mylan v. Sanofi
`IPR2018-01676
`
`