Attributes Influencing Insulin Pen Preference Among Caregivers and Patients With Diabetes Who Require Greater Than 20 Units of Mealtime Insulin

Abstract

Background:

This study compared patient preference for Humalog® KwikPen™ 200 units/mL (insulin lispro; hereafter, IL 200 pen; Eli Lilly and Company, Indianapolis, IN) versus the Humalog KwikPen 100 units/mL (insulin lispro; hereafter, IL 100 pen; Eli Lilly and Company, Indianapolis, IN) in patients with diabetes requiring >20 units of mealtime insulin and diabetes caregivers. This study also determined which attributes had the greatest influence on pen preference selection.

Methods:

In this 2-period, crossover, simulated-use study, 106 participants were randomized to 1 of 8 sequences that varied the pen order (IL 100 pen or IL 200 pen) and dosing order (15 units = low dose or 50 units = high dose) for a total of 4 simulated injections. Participants then completed a self-administered questionnaire to select their overall preference between the 2 pens and then rated the importance of 11 pen attributes in contributing to their overall preference.

Results:

Of the 90 participants expressing an overall preference, significantly more preferred the IL 200 pen to the IL 100 pen (IL 200 pen: 80 respondents; IL 100 pen: 10 respondents; 95% confidence interval [0.81, 0.94], P < .0001). The total amount of insulin in the pen, the ease in pressing the injection button, and the amount of fluid injected were key attributes influencing IL 200 pen preference.

Conclusions:

Based on these key attributes, the IL 200 pen was significantly preferred over the IL 100 pen by patients with diabetes who require >20 daily mealtime insulin units or diabetes caregivers and may improve the injection experience for these patients.

Keywords

basal-bolus therapy human factor studies injection volume insulin pen concentrated insulin diabetes mellitus

As of 2012 in the United States (US), an estimated 21.0 million people have been diagnosed with diabetes, with type 2 diabetes mellitus (T2DM) accounting for approximately 90%-95% of those diagnoses in adults.¹ As of 2014, the estimated worldwide prevalence of diabetes was 387 million (8.3%) people² with the majority of diagnoses for T2DM.³ Because T2DM is a progressive disease, patients may eventually need basal-bolus therapy to achieve glycemic targets when lifestyle modifications or other glucose-lowering therapies are no longer effective.⁴ In addition, some patients with T2DM may have higher insulin requirements particularly if they are overweight or obese.^5
-7 Excess body weight is known to be a major contributor to insulin resistance,⁸ and as insulin resistance worsens, exogenous insulin requirements increase.⁹ Published real-life data from retrospective,^10
-12 prospective observational,^13,14 and noninterventional studies¹⁵ have identified patients with T2DM who report using more than 20 units of mealtime insulin per day. Likewise, pediatric¹⁶ and adult patients with type 1 diabetes mellitus (T1DM)^17,18 have also reported using more than 20 units of mealtime insulin per day. As of 2012, more than 60% of all US patients with T2DM injecting mealtime insulin were taking more than 20 units of their mealtime insulin per day (mealtime defined as rapid-/short-acting insulin; excludes pumpers); in Western Europe (EU5), the percentage was even higher (70% of all mealtime insulin-injecting patients with T2DM were taking 20 or more units of mealtime insulin per day).¹⁹

In response to increasing insulin requirements, concentrated insulins are currently being developed to address patient needs.²⁰ Recently developed concentrated insulins include insulin degludec (200 unit/mL formulation; Novo Nordisk, Bagsværd, Denmark),²¹ insulin glargine (300 unit/mL formulation; sanofi-aventis US LLC, Bridgewater, NJ, USA),²² and insulin lispro (200 unit/mL; Eli Lilly and Company, Indianapolis, IN, USA).²³ The prefilled insulin delivery devices that contain these concentrated insulins dial and dose in actual units such that no conversion is needed based on the concentration.^21
-23

The Humalog® 200 units/mL KwikPenTM.(insulin lispro 200 units/mL, hereafter, IL 200 pen; Eli Lilly and Company, Indianapolis, IN, USA) insulin delivery device was recently approved in the US²⁴ and in the European Union.²⁵ The IL 200 pen insulin delivery device delivers the same unit dose of insulin lispro (200 units/mL) in half the volume as the Humalog® 100 units/mL KwikPenTM (insulin lispro100 units/mL; hereafter, IL 100 pen; Eli Lilly and Company, Indianapolis, IN, USA).²³ Because the IL 200 pen contains 600 units of insulin as compared to the IL 100 pen with 300 units, patients can use the IL 200 pen for twice as many doses of insulin.²⁵ In addition, the glide force, or force to push the injection button, of the IL 200 pen is significantly lower than that of the IL 100 pen for the same dose.²³ The pens are designed to function similarly to facilitate patient ease of use. Both pens are primed in 2-unit increments, with the IL 200 pen priming in half the volume of the IL 100 pen. The purpose of this study was to examine which insulin device delivery option (IL 100 pen or IL 200 pen) was preferred among the intended user population of patients with diabetes who require greater than 20 units of mealtime insulin as well as diabetes caregivers. This study also determined which attributes had the greatest influence on pen preference selection.

Methods

Study design

This single-visit, 2-period, crossover, simulated-use preference study (Figure 1) was conducted over a 3-week period across multiple clinical trial sites in the US in accordance with the Declaration of Helsinki and the International Conference on Harmonisation Guideline on Good Clinical Practices.²⁶ The protocol and informed consent form were reviewed and approved by an institutional review board. Data in this study were collected in accordance with industry best practices, regulatory guidances, and consensus standards that are considered reliable and accurate for human factor studies.

Figure 1.

Study design. IL 100 pen, insulin lispro 100 units/mL prefilled pen; IL 200 pen, insulin lispro 200 units/mL prefilled pen.

Study Participants

Participants were recruited by means of advertisements. Interested participants were prescreened by telephone to determine if they met study eligibility criteria. Adult patients at least 18 years old with T1DM or T2DM who self-administered, either by vial and syringe or injection pen, more than 20 units of mealtime (short- or rapid-acting) insulin per day were included in the study. Patients were excluded if they were physically unable to perform requested tests as determined during screening or only used an insulin pump to deliver insulin instead of a vial and syringe or pen.

Study Procedures

After informed consent was obtained, participants were randomly assigned to 1 of 8 sequences that varied the pen order (IL 100 pen or IL 200 pen) and the dosing order (15 units = low dose or 50 units = high dose) for a total of 4 simulated injections of saline (Figure 1). Simulated injections involved asking the participant to prepare the IL 100 pen or the IL 200 pen, as applicable, and inject the appropriate dose into the injection pad.

Eligible participants received training on the first pen, including information about the insulin and the steps required to administer an injection (eg, attaching a needle, priming the pen, dialing a dose, and injecting), before the respective pen and instructions for use were provided to the participant. Participants then completed 2 sequential injections in a randomized order. The injection task began with the participant attaching a new needle to the pen, priming the pen, dialing the appropriate dose (15 or 50 units), and performing the injection into the injection pad and ended with the participant discarding the needle. After completing both injection tasks with the first pen, participants received training on the second pen and then completed 2 sequential injections of the same unit doses as the first pen in a randomized order. After finishing the 4 simulated injections, participants completed a self-administered questionnaire to evaluate their final preference between the 2 pens.

The questionnaire used in this study was adapted from existing insulin device assessment and preference questionnaires. It was modified specifically for this study to evaluate overall preference for the injection device and then evaluate aspects of the participant’s device experience and specific features of the device that contributed to overall device preference. This questionnaire was adapted from items within the Insulin Device Assessment Battery and the Insulin Device Preference Battery, which were developed as patient-completed tools to compare the characteristics of and patient preference for different insulin delivery devices.^27
-29 Content validity has been established via cognitive debriefing of the original questionnaires.²⁷ The modified measures underwent cognitive debriefing testing in a pilot study with 4 participants to ensure comprehensiveness and comprehension of the preference question and attribute ratings in a population of patients who were using insulin.

The first section of the preference questionnaire evaluated overall preference with 3 response options (prefer IL 100 pen, prefer IL 200 pen, and no preference). The second section of the preference questionnaire presented participants with a list of 11 device features or attributes (Table 1) and asked them to rate the importance of these attributes in choosing their overall preference on a 5-point Likert-type scale where 1 corresponded to “not at all important” and 5 corresponded to “extremely important.”

Table 1.

Reasons for Pen Preference.

Pen attribute
Easy to learn to use the pen
Easy to follow instructions when using the pen
Overall easy to use the pen
Easy to read the dose
Easy to press the button to inject
Color of pen
Total amount of insulin in the pen (number of units)
Amount of liquid injected
Shape of pen
Easy to know I have selected the intended dose for injections
Easy to prime the pen

Adapted from items within the Insulin Device Assessment Battery and the Insulin Device Preference Battery.^27
-29

Statistical Analyses

A sample size of 114 participants was selected based on a 15% dropout rate with 97 participants needed to complete the study. Assuming a 33% rate of no preference, and assuming among those with a preference that the actual rate of preferring the IL 200 pen over the IL 100 pen is 70%, a minimum of 65 participants with preference were needed to show preference for the IL 200 pen over the IL 100 pen at 91% power (ie, demonstrate the primary objective). If the lower limit of an exact 95% 2-sided confidence interval (CI) for preference rate exceeded 50%, then the primary objective was met.

Baseline characteristics were summarized using descriptive statistics (frequencies, means, and standard deviations [SDs]). For the primary objective, descriptive statistics and a 95% CI were used to summarize the preference rate. The Wilson (score) CI for binomial proportion was used to test the null hypothesis (ie, preference rate ≤50%) for the primary objective. Only the primary objective was reported for all participants. Logistic regression models were used to test the effects of sequence (pen) and baseline characteristics on preference as sensitivity analyses.

For the secondary objectives, each question/item was summarized using descriptive statistics for the score (number of subjects, mean, SD, median, first quartile, third quartile, minimum, and maximum) and frequencies and percentages of response for each of the 5 response options. Ranking of attribute importance was based on the percentage of response for each of the 5 response options as well as the mean score. A mixed-model repeated measures analysis was used to compare the scores for the top 3 and top 5 attributes influencing pen preference, respectively. Fixed effects included the corresponding questions, sequence (pen), and baseline characteristics. Within-subject covariance structure was included in the model to account for the repeated-measure nature of the data (scores for different questions were related within subject). The adjustment for multiple testing (Tukey) was used to perform comparisons of scores between the top 3 or top 5 attributes influencing pen preference.

Results

Patient Disposition

A total of 106 participants signed and dated the informed consent form and were randomized to 1 of 8 sequences. Of those 106 randomized participants, 5 (4.7%) were discontinued from the study because it was determined that they were not self-administering more than 20 units of mealtime insulin per day (including any participants who were injecting an insulin mixture). The remaining 101 (95.3%) participants completed all 4 injections.

Patient Characteristics and Demographics of All Completers

Participants in this study had a mean age of 54 years, with similar numbers of males and females, and were primarily white (Table 2). The majority of patients had been diagnosed with T2DM, had a mean duration of diabetes of 17 years, and self-administered their insulin injections. Baseline characteristics of the patients who completed the study were similar across the 3 groups (prefer IL 100 pen, prefer IL 200 pen, and no preference).

Table 2.

Baseline Demographics for All Participants.

Variable	Total (N = 101)	P value^a
Injection giver, n
Caregiver	5	.4769
Patient	96
Gender, n
Female	45	.7677
Male	56
Age (years), mean ± SD	54 ± 14	.3711
Race, n
American Indian	1	.9703
Asian	1
Black	20
Hispanic	2
>1 race	1
White	76
Education level, n
College graduate	43	.1743
High school graduate	11
Postgraduate	15
Some college	27
Some high school	2
Some postgraduate	3
Duration of diabetes (years), mean ± SD	17 ± 12	.5773
Diabetes type
Type 1 diabetes	24	.4265
Type 2 diabetes	72
N/A^b	5
Number of daily mealtime units, mean ± SD	54 ± 39	.2360
≥1 meal over 20 units, n	36	.0715

IL 100 pen, insulin lispro 100 units/mL prefilled pen; IL 200 pen, insulin lispro 200 units/mL prefilled pen; n, number in specified category; SD, standard deviation.

P values are for comparisons across the 3 groups (prefer IL 100 pen, prefer IL 200 pen, no preference).

Caregivers did not report the duration of diabetes and reported diabetes type as N/A.

Primary Objective

Of the 90 participants expressing a preference, significantly more participants preferred the IL 200 pen than the IL 100 pen (Figure 2) (IL 200 pen: 80 participants; IL 100 pen 10 participants; 95% CI [0.81, 0.94], P < .0001), meeting the primary objective because the lower limit of the 95% CI exceeded 50%. The remaining 11 participants exhibited no preference for either insulin pen device. Sensitivity analyses of overall preference showed no significant effects of baseline characteristics (eg, age, race, duration of diabetes, sex, mealtime units, type of diabetes) on pen preference (data not shown). In addition, there was no effect of randomization sequence on pen preference.

Figure 2.

Overall preference for the insulin lispro 200 units/mL prefilled pen among respondents declaring a preference. IL 100 pen, insulin lispro 100 units/mL prefilled pen; IL 200 pen, insulin lispro 200 units/mL prefilled pen.

Secondary Objectives

Among the 90 participants who exhibited a preference for either pen device, mean scores were calculated and used to rank all 11 questionnaire items from highest mean score (5 = “extremely important”) to lowest mean score (1 = “not at all important”). The most important attributes influencing preference for IL 200 pen among the 80 participants were the total amount of insulin in the pen (#1) (mean ± SD: 4.16 ± 1.34), the injection button was easy to press (#2) (3.78 ± 1.64), the amount of fluid injected (#3) (3.61 ± 1.61), the overall ease in using the pen (#4) (3.25 ± 1.73), and the ease in reading the dose (#5) (2.90 ± 1.84). The shape (#10) (1.95 ± 1.43) and the color (#11) (1.71 ± 1.30) of the pen were the least important attributes (Figure 3).

Figure 3.

Summary of reasons for IL 200 pen preference by ranking. Data are presented as means and standard deviations. N = 80. *p = .0104 for injection amount vs insulin amount; ^#P = .0299 for amount of fluid injected vs total amount of insulin in pen; ^†P = .0466 for amount of fluid injected vs ease in reading dose; ^‡P = .0016 for total amount of insulin amount in pen vs overall ease in using pen; ^§P < .0001 for total amount of insulin in pen vs ease in reading dose; ^||P = .0062 for overall ease in using pen vs ease in pressing injection button; ^¶P = .0377 for overall ease in use of pen vs ease in reading dose; **P = .0001 for ease in pressing injection button vs ease in reading dose. IL 200 pen, insulin lispro 200 units/mL prefilled pen.

The most important attributes influencing preference for the IL 100 pen among the 10 participants were the injection button was easy to press (#1) (3.70 ± 1.77), the ease in knowing the intended dose (#2) (3.30 ± 2.00), the ease in reading the dose (#3) (3.20 ± 1.62), the overall ease in using the pen (#4) (3.10 ± 1.85), and total amount of insulin in the pen (#5) (2.90 ± 1.45). The shape (#10) (2.50 ± 1.78) and color (#11) (1.80 ± 1.32) of the pen were rated as the least important attributes (Figure 4).

Figure 4.

Summary of reasons for IL 100 pen preference by ranking. Data are presented as means and standard deviations. N = 10. IL 100 pen, insulin lispro 100 units/mL prefilled pen.

Comparison of the top 3 attributes influencing preference for the IL 200 pen based on the mean score depicted in Figure 3 showed a significant difference overall (Table 3, P = .0141). Pairwise comparisons showed a significant difference between the amount of fluid injected (#3) and the total amount of insulin in the pen (#1) (P = .0104) but not for the remaining pairwise comparisons. Comparison of the top 5 attributes influencing preference for the IL 200 pen based on the mean score showed a significant overall difference (P < .0001). Pairwise comparisons of these top 5 attributes were statistically significant, except for the amount of fluid injected (#3) and the overall ease in using the pen (#4), the amount of fluid injected (#3) and the injection button was easy to press (#2), and the total amount of insulin in the pen (#1) and the injection button was easy to press (#2).

Table 3.

Summary and Analysis of the Top 3 and Top 5 Attributes Influencing IL 200 Preference by Ranking.

		LS means difference ± SE	Tukey P value	Overall P value
Pairwise comparison of the top 3 attributes based on mean score
Amount of fluid injected	Total amount of insulin in pen	−0.5500 ± 0.1822	.0104	.0141
Amount of fluid injected	Easy to press button to inject	−0.1625 ± 0.2409	.7792
Total amount of insulin in pen	Easy to press button to inject	0.3875 ± 0.2519	.2809
Pairwise comparison of the top 5 attributes based on mean score
Amount of fluid injected	Total amount of insulin in pen	−0.5500 ± 0.1822	.0299	<.0001
Amount of fluid injected	Overall easy to use pen	0.3625 ± 0.2375	.5500
Amount of fluid injected	Easy to press button to inject	−0.1625 ± 0.2409	.9611
Amount of fluid injected	Easy to read dose	0.7125 ± 0.2504	.0466
Total amount of insulin in pen	Overall easy to use pen	0.9125 ± 0.2273	.0016
Total amount of insulin in pen	Easy to press button to inject	0.3875 ± 0.2519	.5424
Total amount of insulin in pen	Easy to read dose	1.2625 ± 0.2372	<.0001
Overall easy to use pen	Easy to press button to inject	−0.5250 ± 0.1467	.0062
Overall easy to use pen	Easy to read dose	0.3500 ± 0.1195	.0377
Easy to press button to inject	Easy to read dose	0.8750 ± 0.1852	.0001

IL 200 pen, insulin lispro 200 units/mL prefilled pen; LS, least squares; MMRM, mixed model repeated measures. P values were calculated from MMRM model with response, sequence, age, sex, race, education level, number of meal time units, at least 1 meal over 20 units (Y/N), and injection giver as covariates and the variance-covariance structure for response specified as unstructured.

LS means used for pairwise differences.

Comparison of the top 3 attributes influencing preference for the IL 100 pen showed no significant overall differences (P = .2267). Similarly, there were no significant overall differences among the top 5 attributes (P = .1735).

In the IL 200 preference group, all questionnaire items were ranked based on the percentages of responses for each of the 5 response options. The ranking of items based on the percentage of participants who felt they were “extremely important” was identical to the mean score ranking. More than 50% of participants ranked the top 2 attributes relating to the IL 200 pen preference as “extremely important” (Figure 5): the total amount of insulin in the pen (#1) (64%) and the injection button was easy to press (#2) (55%).

Figure 5.

Summary of percentage of reasons for IL 200 pen preference. IL 200 pen, insulin lispro 200 units/mL prefilled pen.

Discussion

Most of the patients with diabetes requiring more than 20 units of mealtime insulin or diabetes caregivers preferred the IL 200 pen over the IL 100 pen, and this result met the primary objective. Key pen attributes influencing preference for the IL 200 pen included the total amount of insulin in the pen, the ease of pressing the injection button, and the amount of fluid injected, which are attributes specific to the higher concentration of insulin in the IL 200 pen.²³ Participants noted anecdotally that they liked the idea of using fewer pens (facilitated by more insulin in each pen), that the pen button was easier to push (due to decreased glide force), and that an injection would be delivered in half the amount of liquid in the IL 200 pen compared to the IL 100 pen (due to the concentration). Questionnaire items that were rated less important by participants who preferred the IL 200 pen do not suggest such attributes were deemed less important in a pen device, but rather did not exert as much influence over the preference decision.

Insulin therapy, particularly injectable insulin therapy, is associated with several unique challenges that may negatively impact patient acceptability of treatment and consequently reduce adherence.³⁰ Patients who use insulin frequently have described how dosing is inflexible, how injections disrupt daily activities, and expressed dissatisfaction with the number of daily injections required.^31
-33 Poor adherence, especially among patients with T2DM, is associated with more frequent dosing,^34,35 complex dosing regimens (eg, number of therapies), and burden of administration.³⁵ Satisfaction with the insulin delivery device (a combined measure of ease of use, interference with daily activities, and social acceptability) has been found to be positively linked to patient self-efficacy in patients with T2DM.³⁶ A positive injection experience that is preferred by patients may lead to improved patient satisfaction, and therefore, better adherence to therapy, which in turn improves longer-term patient outcomes. Patients have indicated they would be more compliant with their insulin therapy if they can use a product to reduce injection-related pain.³¹ Patients with diabetes who are more satisfied with treatment may be more likely to adhere to it,³⁷ so it is important to individualize treatment according to patient preferences so that treatment satisfaction is maximized. Because the IL 200 pen contains an increased strength formulation of insulin lispro where each dose from the IL 200 pen is delivered in half the volume of the IL 100 pen, the pen’s glide force is reduced²³ and there are fewer pen transitions for the user. These benefits may help address the challenges associated with poor treatment adherence.

Although this study assessed patient preferences among intended users of the insulin pen devices, it was a simulated injection study limited to a single visit in a setting intended to mimic the environment where injections occur (ie, home, community settings, outpatient facilities). However, injections may also occur in other environments, such as restaurants or places of work. Because this study used simulated injections, we could not assess whether participants would perceive actual injections of the same unit dose differently (eg, if a higher concentration insulin injection would be less painful), which could influence preference. Further study is warranted to assess real-world injection. Although our study recruited participants with prior pen or vial and syringe experience, there was not enough statistical power to conduct a subanalysis on these patient populations; therefore, we cannot conclude whether there was a differential pattern of pen preference based on the prestudy use of a pen versus a syringe and vial. The questionnaire was not a validated instrument; however, cognitive debriefing testing of these measures was done in the intended population of patients who were using insulin to ensure the breadth and understanding of the preference question and attribute ratings.

Conclusions

In summary, patients with diabetes who require more than 20 daily mealtime insulin units or diabetes caregivers preferred the IL 200 pen to the IL 100 pen based on the total amount of insulin in the pen device, the ease of pressing the injection button, and amount of fluid injected. These findings suggest that a more concentrated insulin pen device may improve the mealtime insulin injection experience for patients using greater than 20 units a day of mealtime insulin.

Footnotes

Acknowledgements

The authors would like to thank Eileen Girten (inVentiv Health Clinical) for assistance with the preparation of this article.

Abbreviations

CI, confidence interval; EU, European Union, IL100 pen, insulin lispro 100 units/mL prefilled pen; IL200 pen, insulin lispro 200 units/mL prefilled pen; SD, standard deviation; T2DM, type 2 diabetes mellitus.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: TW, KAC, KVB, and TMR are employees of Eli Lilly and Company. TW, KVB, and TMR are shareholders for Eli Lilly and Company.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by Eli Lilly and Company.

References

Centers for Disease Control and Prevention. National Diabetes Statistics Report, 2014. Available at: http://www.cdc.gov/diabetes/pubs/statsreport14/national-diabetes-report-web.pdf. Accessed August 18, 2015.

International Diabetes Foundation. IDF Diabetes Atlas Poster 2014 Update. Available at: https://www.idf.org/sites/default/files/Atlas-poster-2014_EN.pdf. Accessed August 18, 2015.

International Diabetes Foundation. IDF Diabetes Atlas, Sixth Edition. 2013. Available at: http://www.idf.org/sites/default/files/EN_6E_Atlas_Full_0.pdf. Accessed August 18, 2015.

Inzucchi

Bergenstal

Buse

. Management of hyperglycemia in type 2 diabetes, 2015: a patient-centered approach: update to a position statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care. 2015;38(1):140-149.

King

Armstrong

. Basal bolus dosing: a clinical experience. Curr Diabetes Rev. 2005;1(2):215-220.

Watson

Wilson

Alsop

Kumar

. Weight and glycaemic control in type 2 diabetes: what is the outcome of insulin initiation? Diabetes Obes Metab. 2011;13(9):823-831.

Rodbard

Gough

Lane

Korsholm

Bretler

Handelsman

. Reduced risk of hypoglycemia with insulin degludec versus insulin glargine in patients with type 2 diabetes requiring high doses of basal insulin: a meta-analysis of 5 randomized BEGIN trials. Endocr Pract. 2014;20(4):285-292.

Kahn

Flier

. Obesity and insulin resistance. J Clin Invest. 2000;106(4):473-481.

Ovalle

. Clinical approach to the patient with diabetes mellitus and very high insulin requirements. Diabetes Res Clin Pract. 2010;90(3):231-242.

10.

Aagren

Luo

Moës

. Healthcare utilization changes in relation to treatment intensification with insulin aspart in patients with type 2 diabetes. Data from a large US managed-care organization. J Med Econ. 2010;13(1):16-22.

11.

Cai

Han

Luo

. Analysis of insulin doses of Chinese type 2 diabetic patients with intensive insulin treatment. PLOS ONE. 2012;7(6):e38962.

12.

Miao

Wei

Baser

Xie

. Real world outcomes of adding rapid-acting insulin versus switching to analog premix insulin among US patients with type 2 diabetes treated with insulin glargine. Patient Prefer Adherence. 2013;7:951-960.

13.

Reaney

Cypryk

Tentolouris

. Resource utilisation and clinical data before and after switching between short-acting human insulin and rapid-acting insulin analogues in patients with type 2 diabetes: the SWING study. Diabetes Res Clin Pract. 2012;97(2):231-241.

14.

Mathieu

Storms

Tits

Veneman

Colin

. Switching from premixed insulin to basal-bolus insulin glargine plus rapid-acting insulin: the ATLANTIC study. Acta Clin Belg. 2013;68(1):28-33.

15.

Bebakar

Lim-Abrahan

Jain

Seah

Soewondo

. Safety and effectiveness of insulin aspart in type 2 diabetic patients: results from the ASEAN cohort of the A1chieve study. Diabetes Res Clin Pract. 2013;100(suppl 1):S17-S23.

16.

Thalange

Deeb

Iotova

. Insulin degludec in combination with bolus insulin aspart is safe and effective in children and adolescents with type 1 diabetes. Pediatr Diabetes. 2015;16(3):164-176.

17.

Davies

Gross

Ono

. Efficacy and safety of insulin degludec given as part of basal-bolus treatment with mealtime insulin aspart in type 1 diabetes: a 26-week randomized, open-label, treat-to-target non-inferiority trial. Diabetes Obes Metab. 2014;16(10):922-930.

18.

Home

Bergenstal

Bolli

. New insulin glargine 300 units/mL versus glargine 100 units/mL in people with type 1 diabetes: a randomized, phase 3a, open-label clinical trial edition (EDITION 4). Diabetes Care. 2015;38(12):2217-2225.

19.

20.

Segal

El Sayed

. Are you ready for more insulin concentrations? J Diabetes Sci Technol. 2015;9(2):331-338.

21.

Gough

SCL

Bhargava

Jain

Meserbach

Rasmussen

Bergenstal

. Low-volume insulin degludec 200 units/mL once daily improves glycemic control similarly to insulin glargine with a low risk of hypoglycemia in insulin-naïve patients with type 2 diabetes: a 26-week randomized, controlled, multinational, treat-to-target trial: the BEGIN LOW VOLUME trial. Diabetes Care. 2013;36(9):2536-2542.

22.

Goldman

White

Jr.

New insulin glargine 300 U/mL for the treatment of type 1 and type 2 diabetes mellitus. Ann Pharmacother. 2015;49(10):1153-1161.

23.

Rees

Lennartz

Ignaut

. A comparison of glide force characteristics between 2 prefilled insulin lispro pens. J Diabetes Sci Technol. 2015;9(2):316-319.

24.

PR Newswire. U.S. Food and Drug Administration approves Humalog® 200 units/mL KwikPen®. Available at: http://www.prnewswire.com/news-releases/us-food-and-drug-administration-approves-humalog-200-unitsml-kwikpen-300089341.html. Accessed August 19, 2015.

25.

PR Newswire. European Commission grants marketing authorization to Humalog® 200 units/ml KwikPen™ insulin. 2014. Available at: http://www.prnewswire.com/news-releases/european-commission-grants-marketing-authorisation-to-humalog-200-unitsml-kwikpen-insulin-277989761.html. Accessed August 19, 2015.

26.

World Medical Association Declaration of Helsinki. World Medical Association Declaration of Helsinki. Recommendations guiding physicians in biomedical research involving human subjects. JAMA. 1997;277(11):925-926.

27.

Ignaut

Schwartz

Sarwat

Murphy

. Comparative device assessments: Humalog KwikPen compared with vial and syringe and FlexPen. Diabetes Educ. 2009;35(5):789-798.

28.

Clark

Valentine

Bodie

Sarwat

. Ease of use and patient preference injection simulation study comparing two prefilled insulin pens. Curr Med Res Opin. 2010;26(7):1745-1753.

29.

Wong

Abdulnabi

. Ease of use of two reusable, half-unit increment dosing insulin pens by adult caregivers of children with type 1 diabetes: a randomized, crossover comparison. J Diabetes Sci Technol. 2013;7(2):582-583.

30.

Aronson

. The role of comfort and discomfort in insulin therapy. Diabetes Technol Ther. 2012;14(8):741-747.

31.

Rubin

Peyrot

Kruger

Travis

. Barriers to insulin injection therapy: patient and health care provider perspectives. Diabetes Educ. 2009;35(6):1014-1022.

32.

Peyrot

Rubin

Kruger

Travis

. Correlates of insulin injection omission. Diabetes Care. 2010;33(2):240-245.

33.

Peyrot

Barnett

Meneghini

Schumm-Draeger

. Insulin adherence behaviours and barriers in the multinational Global Attitudes of Patients and Physicians in Insulin Therapy study. Diabet Med. 2012;29(5):682-689.

34.

Ingersoll

Cohen

. The impact of medication regimen factors on adherence to chronic treatment: a review of literature. J Behav Med. 2008;31(3):213-224.

35.

García-Pérez

Álvarez

Dilla

Gil-Guillén

Orozco-Beltrán

. Adherence to therapies in patients with type 2 diabetes. Diabetes Ther. 2013;4(2):175-194.

36.

Hayes

Curtis

Ilag

Nelson

Wong

Funnell

. Expectations about insulin therapy, perceived insulin-delivery system social acceptability, and insulin treatment satisfaction contribute to decreases in insulin therapy self-efficacy in patients with type 2 diabetes after 36 weeks insulin therapy. J Diabetes. 2013;5(3):358-367.

37.

Davies

Speight

. Patient-reported outcomes in trials of incretin-based therapies in patients with type 2 diabetes mellitus. Diabetes Obes Metab. 2012;14(10):882-892.