Scholarly article on topic 'Behaviours, thoughts and perceptions around mealtime insulin usage and wastage among people with type 1 and type 2 diabetes mellitus: a cross-sectional survey study'

Behaviours, thoughts and perceptions around mealtime insulin usage and wastage among people with type 1 and type 2 diabetes mellitus: a cross-sectional survey study Academic research paper on "Economics and business"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{"Mealtime insulin" / "Rapid-acting insulin" / "Type 2 diabetes" / "Type 1 diabetes" / Wastage / "Patient-reported outcomes"}

Abstract of research paper on Economics and business, author of scientific article — Kate Van Brunt, Riccardo Pedersini, Jillian Rooney, Sheila M. Corrigan

Abstract Aims People with diabetes who use mealtime insulin (MTI) were surveyed about insulin wastage and injection habits when insufficient insulin remains in a disposable prefilled pen/cartridge to administer a full dose in a single injection. Methods Cross-sectional, online, self-reported survey of MTI usage/wastage behaviour in 400 adults with type 1 (n =120) or type 2 (n =280) diabetes mellitus administering >20units/day of MTI via 100units/ml prefilled pens/cartridges for ⩾1month, conducted in France, Germany, Italy and UK. Results Participants’ mean±standard deviation age was 54.5±12.2years, body mass index was 29.9±7.2kg/m2 and duration of MTI therapy was 8.6±7.8years. They administered 3.7±5.9 injections/day with meals, using 11.3±18.0 prefilled pens/cartridges per month. Overall, 63.5% split the dose across two prefilled pens/cartridges (i.e. administered two injections to obtain a full dose), 15.0% used just what remained in their current pen (i.e. took a lower-than-prescribed dose) and 36.3% discarded prefilled pens/cartridges still containing insulin (i.e. took full dose with new pen). The latter participants discarded a mean 5.5±8.2 prefilled pens/cartridges monthly still containing insulin, each containing 8.6±8.7 units of insulin. Participants who wasted insulin considered it frustrating, time-consuming and painful to inject twice. Conclusions Patients taking >20units/day MTI can find transitions between insulin pens challenging. This study highlights the need to identify ways of improving transitions between pens to make transitions easier for insulin users, which could potentially improve adherence to prescribed doses and reduce waste.

Academic research paper on topic "Behaviours, thoughts and perceptions around mealtime insulin usage and wastage among people with type 1 and type 2 diabetes mellitus: a cross-sectional survey study"

DIABETES

RESEARCH and CLINICAL PRACTICE

Accepted Manuscript

Behaviours, thoughts and perceptions around mealtime insulin usage and wastage among people with type 1 and type 2 diabetes mellitus: a cross-sectional survey study

Kate Van Brunt, Riccardo Pedersini, Jillian Rooney, Sheila M. Corrigan

PII: DOI:

Reference: To appear in:

S0168-8227(16)31748-X http://dx.doi.org/10.1016/j.diabres.2016.12.002 DIAB 6819

Diabetes Research and Clinical Practice

Received Date: Revised Date: Accepted Date:

7 April 2016 16 November 2016 5 December 2016

Please cite this article as: K. Van Brunt, R. Pedersini, J. Rooney, S.M. Corrigan, Behaviours, thoughts and perceptions around mealtime insulin usage and wastage among people with type 1 and type 2 diabetes mellitus: a cross-sectional survey study, Diabetes Research and Clinical Practice (2016), doi: http://dx.doi.org/10.1016/ j.diabres.2016.12.002

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Behaviours, thoughts and perceptions around mealtime insulin usage and wastage among people with type 1 and type 2 diabetes mellitus: a cross-sectional survey study

Author(s):

Author affiliation(s):

tooneyd ar

Kate Van Brunta, Riccardo Pedersinib'c, Jillian Rooneya and Sheila M Corrigan9 aEli Lilly & Company Ltd, Erl Wood Manor, Sunninghill Road, Windlesham, Surrey GU20 6PH, UK; van_brunt_kate@lilly.com, bKantar Health, The Kirkgate, 19-31 Church Street, Epsom, Surrey KT17 4PF, UK; cRTI Health Solutions, Travessera de Gracia 56, Ático 1a, 08006 Barcelona, Spain; riccardo.pedersini@gmail.com, dKantar Health, 11 Madison Avenue 12th Floor, New York, NY 10010, USA; Jillian.Rooney@kantarhealth.com, and eEli Lilly & Company, Lilly Corporate Center, Indianapolis, Indiana 46285, USA; corrigan_sheila_m@lilly.com

Corresponding author: Kate Van Brunt Address: ¿> ' Eli Lilly & Company Ltd

Phone: Email:

Erl Wood Manor, Sunninghill Road, Windlesham Surrey GU20 6PH UK

+44 (0)7712888062 van_brunt_kate@lilly.com

Running header:

Mealtime insulin wastage

Abstract

Aims: People with diabetes who use mealtime insulin (MTI) were surveyed about insulin wastage and injection habits when insufficient insulin remains in a disposable prefilled pen/cartridge to administer a full dose in a single injection.

Methods: Cross-sectional, online, self-reported survey of MTI usage/wastage behaviour in 400 adults with type 1 (n=120) or type 2 (n=280) diabetes mellitus administering >20 units/day of MTI via 100 units/ml prefilled pens/cartri month, conducted in France, Germany, Italy and UK.

llitus ridges for >1

± 12.2 vf

Results: Participants' mean ± standard deviation age was 54.5 ± 12.2 years, body mass index was 29.9 ± 7.2 kg/m2 and duration of MTI therapy was 8.6 ± 7.8 years. They administered 3.7 ± 5.9 injections/day with meals, using 11.3 ± 18.0 prefilled pens/cartridges per month. Overall, 63.5% split the dose across two prefilled pens/cartridges (i.e. administered two injections to obtain a full dose), 15.0% used just what remained in their current pen (i.e. took a lower-than-prescribed dose) and 36.3% discarded prefilled pens/cartridges still containing insulin (i.e. took full dose with new pen). The latter participants discarded a mean 5.5 ± 8.2 prefilled pens/cartridges monthly still containing insulin, each containing 8.6 ± 8.7 units of insulin. Participants who wasted insulin considered it frustrating, time-consuming and painful to inject twice.

Conclusions: Patients taking >20 units/day MTI can find transitions between insulin pens challenging. This study highlights the need to identify ways of improving transitions between pens to make transitions easier for insulin users, which could potentially improve adherence to prescribed doses and reduce waste.

Keywi wasta

ords: Mealtime insulin, rapid-acting insulin, type 2 diabetes, type 1 diabetes, tage, patient-reported outcomes

1. Introduction

Insulin is the cornerstone of treatment for people with type 1 diabetes mellitus [1 ], and many people with type 2 diabetes mellitus will eventually need insulin therapy to help maintain glycaemic control [2]. Therapy with insulin may comprise a combination o basal and mealtime insulin [1,2]. Doses of total insulin, including mealtime insulin, are tending to increase due to greater emphasis on the optimisation of therapy [1] and because of the worldwide trend towards increased bodyweight [3]. People with diabetes who have insulin resistance and/or a higher body mass index (BMI) need more daily insulin than people without insulin resistance or with a lower BMI respectively [4,5].

to giv

Mealtime insulin, in the form of rapid/short-acting insulin, is often administered using disposable prefilled pens or reusable pens with disposable cartridges [6,7]. People with diabetes who have higher insulin requirements will use a greater number of pens/cartridges. Moreover, prefilled pens/cartridges will generally not contain an exact multiple of the dose required, and people may behave in various ways as they approach the change from a used to a new prefilled pen/cartridge. Instructions to people with diabetes typically guide them to follow one of two scenarios when a pen/cartridge has insufficient insulin remaining to provide a full dose [8]. The first option is to give a partial dose using the current pen and the remainder of the dose using a pen, which means the person must administer two injections. The second option is give the full dose with a new pen; this involves only a single injection but means that some unused insulin remains in the discarded pen and is therefore wasted. A third, non-recommended, option is also possible: some people may decide to administer just the insulin remaining in their current pen, and therefore receive less than the prescribed dose. Individuals might do this to avoid a second injection or because they do not have a second pen with them.

Each of these behaviours has potential implications for healthcare systems and/or the individual being treated. For example, people with diabetes may find administering two insulin injections to complete one dose is a burden, which may affect treatment adherence and health-related quality of life (HRQoL). There are inverse relationships

between number of injections and adherence [9] or HRQoL [10] among people with diabetes. Moreover, taking less insulin than prescribed may result in day-to-day glucose variability and could potentially have an adverse effect on glycaemic control [9,11]. Wastage of insulin may have cost implications for healthcare systems, which may be significant for payers, given increasingly constrained healthcare budgets and rising healthcare costs.

However, very little information is available about which of the possible behaviours people with diabetes are most likely to display when they reach the end of a prefilled pen/cartridge, or about their reasons for choosing a particular option. A search of the medical literature identified only one study on this topic. This single-centre study in the United Kingdom (UK) dealt specifically with insulin wastage and found that one-quarter of people with diabetes were so concerned about whether or not they wasted insulin when they approached a change from a used to a new pen/cartridge that they administered two injections to avoid any wastage [12]. The participants in this study from 1997 were using premixed insulin (30% short-acting/70% intermediate-acting), and the results may not be generalisable to those using mealtime insulin, because actions of fixed insulin mixtures contain a larger dose than is used for injections taining rapid/short-acting insulin alone. Steel et al. noted that less insulin was wasted when using a 300-unit disposable than using a 150-unit cartridge [12].

inject conta

Individuals using pens that contain more units per pen are not faced with the decision about how to handle an empty pen as often as individuals using pens that contain fewer units per pen. Today, due to the availability of new higher strength (200 U/mL) mealtime insulin, prefilled pens that contain 600 units are now an alternative to

standard 300 unit (100 U/mL) pens, and may provide a potential advantage for patients [13,14]. Since there are scant data, and no recent research, about this aspect of the management of diabetes, a survey study was conducted to better understand how mealtime insulin users behave when they do not have enough insulin remaining in a

prefilled pen/cartridge to complete a full dose with one injection. We selected insulin wastage as our main measure because the results of the Steele et al. study [12] allowed us to generate hypotheses around this outcome. Our survey additionally allowed us to explore the impact of insulin wastage on HRQoL and better understand the behaviours that mealtime insulin users, including those who waste insulin, display when changing from a used to a new insulin pen. The primary objective was therefore to estimate the amount of mealtime insulin that is wasted by people with type 1 or type 2 diabetes based on their self-reported injection habits. Secondary objectives were to understand participants' perceptions and attitudes towards their mealtime insulin use and to determine whether there was any association between insulin wastage and HRQoL.

ny asso /

2. Subjects and methods

This was a cross-sectional study in which people with type 1 or type 2 diabetes mellitus from France, Germany, Italy and the UK, completed an online survey during February-March 2015. The survey was administered to 400 participants (100 per country, comprising 30 with type 1 diabetes mellitus and 70 with type 2 diabetes mellitus)

illitus).

Panel, a

Participants were recruited from among the panel members at Lightspe professional survey recruitment company. The study used one of the company's online research panels; individuals have to opt into the panel, and must have access to a computer or internet device. Invitations to access an online survey link were sent to all panel members who had type 1 or type 2 diabetes mellitus. Only those respondents who provided informed consent and met the study criteria on the screening page could proceed to access the survey. Enrolment into the study continued until the pre-planned target of 400 participants (120 type 1 diabetes mellitus and 280 type 2 diabetes mellitus) had completed the survey. All participants provided informed consent.

2.1. Study sample

The study included people aged at least 18 years with type 1 or type 2 diabetes mellitus that had been diagnosed by a healthcare provider, who had been taking more than 20 units of mealtime insulin per day using a 100 unit/ml disposable prefilled pen or reusable pen with disposable cartridge to administer their rapid/short-acting insulin for more

re than 1 month, and who were able to understand the survey in their local language.

Respondents were excluded if they only took mixed/biphasic, intermediate- or long-acting insulin, or if they administered mealtime insulin using a syringe with insulin vial or insulin pump.

2.2. Survey instrument

The survey, which took approximately 20 minutes to complete, incorporated the

following items:

• Questions on sociodemographic and health characteristics, several of which A (glycated haemoglobin [HbA1c], exercise, weight, income, education and

employment status) were derived from Kantar Health's National Health and Wellness Survey, a global self-reported general population survey that provides disease-specific data for the healthcare sector and includes data from five European Union countries - France, Germany, Italy, Spain and the UK [15].

• A newly developed set of questions designed to understand insulin usage and wastage behaviour of people with diabetes taking mealtime insulin, including general usage information (e.g. length of time on insulin therapy, the average number of prefilled pens and units of insulin used), behaviour when the end of a prefilled pen/cartridge was reached (e.g. use of dual injections, administration of smaller-than-recommended doses, and discarding prefilled pens still containing insulin) and attitudes specifically towards mealtime insulin use. The questionnaire is available as online supplementary material. These questions were newly developed because a validated instrument to assess mealtime insulin usage and behaviours was not available. Respondents were asked separate questions about each of the

C behaviours of interest related to pen transitions, and could report practising all that applied to them. They were then asked about the frequency of each behaviour they ^^^^ reported practising. The number of times two injections were administered for a

single dose, or a smaller-than-prescribed dose of insulin was taken, was based on a 3-month period; the number of times that pens were discarded still containing insulin was based on a 1-month period; additional details on recall periods are provided in the questionnaire (online supplementary material).

Validated patient-reported outcome questionnaires, including the EuroQol 5-Dimension 5 level (EQ-5D-5L) to assess HRQoL [16] and the Perceptions of Insulin Therapy Questionnaire (PITQ, formerly the Experience with Insulin Therapy Questionnaire), which incorporates the Self-Efficacy about Insulin Therapy Questionnaire (SEITQ), to evaluate participants' attitudes, behaviour and confidence regarding their insulin usage [17,18,19,20]. Brief descriptions of these instruments are provided as online supplementary material.

To ensure the final survey instrument was understood by participants and that it provided comprehensive coverage of concepts relevant to their experiences, ten pilot interviews were performed with people with type 1 or type 2 diabetes mellitus in the UK who were using mealtime insulin. These interviews were conducted via telephone using a computer sharing system to display the English version of the initial questionnaire to the participants. Interviewees were asked to mention any items they felt were unclear or confusing in terms of language, readability or comprehensibility, and were also asked to mention any additional topics they felt should be included. The final survey incorporated minor refinements to the newly developed set of questions based on feedback from these interviews. No changes were made to validated instruments.

k from the

Q-5D-5L were beha

The final questionnaire was translated into local languages. For validated instruments (EQ-5D-5L and PITQ), linguistically validated translations provided by the scale owners

used. The remaining survey questions pertaining to insulin usage, wastage ehaviour and attitudes, were translated using a standard translation process by Transperfect. This company translated the questionnaire (using proprietary software and a team of native linguists) and used feedback from a second company (Kantar Health who used local affiliates in the relevant countries to review draft translations) to produce the final translation.

popu set o

2.3. Statistical methods

A convenience sample of 100 completed participants per country was used for the study. This was the maximum sample size that was expected to be achievable, given the study inclusion/exclusion criteria. A power calculation was not performed because of the lack of prior literature to inform such a calculation. Although the study by Steel et al. [12] provided an estimate of the mean wastage per year among people witf diabetes who were self-administering insulin, it did not report an index of dispersion that could be used to help derive a sample size for this study. The ratio of 30 participants with type 1 diabetes to 70 with type 2 diabetes for each country was chosen to reflect the fact that type 2 diabetes is more prevalent than type 1 diabetes [21], while still providing a sufficient sample size of participants with type 1 diabetes per country to allow for meaningful conclusions to be drawn (the minimum set size for applying the law of large numbers to a subsample is 30 [22]).

All participants who fulfilled the study criteria, completed the survey and provided consent to release information were included in the analyses. There were no missing data in the study.

Statistical analyses were conducted using STATA statistical software (version 13). The primary endpoint, insulin wastage, was summarised using descriptive statistics, including mean values ± standard deviation (SD), and also range for the total pulation, and was based on responses to a single item within the newly developed f questions. In addition, because outliers were observed in the data, insulin wastage variables were bootstrapped (which involves random re-sampling with replacement), using 1,000 repetitions, to obtain robust confidence intervals, with the final result summarised as a mean value ± standard error (SE) and 95% confidence interval (95% CI). Sociodemographic and health characteristics were also summarised using descriptive statistics, including frequencies and percentages for categorical variables, and means ± SD for cardinal variables.

Two sets of bivariate analyses were then performed, one comparing participants who wasted no insulin with participants who wasted insulin, and the other comparing people with type 1 diabetes mellitus with those with type 2 diabetes mellitus (with respect to wastage and other characteristics). Chi-squared tests were used for categorical

variables and t-tests for cardinal variables.

Secondary endpoints, including attitudes towards insulin therapy (PITQ), attitudes

towards mealtime insulin and wastage (newly developed questions) and HRQoL (EQ-5D) were summarised using descriptive statistics. Items within the newly-developed survey were scored individually. Bivariate analyses were conducted as described for the primary endpoint, for the PITQ total scores, and for the EQ-5D total scale score and individual dimensions. Online supplementary material includes the complete survey, which provides further details on the exact question wording, response options and recall period for each survey item.

3. Results

The disposition of participants is summarised in Fig. 1. Among 18,366 members of the Lightspeed Panel who were sent invitations to participate, 7,845 chose to access the online link to the survey site, of whom 7,503 provided informed consent and completed the online screening page, which assessed their eligibility to participate. Among the respondents who were screened, 7,058 did not meet the inclusion/exclusion criteria. Among the 445 respondents who met the inclusion/exclusion criteria, 400 proceeded to complete the survey. The remaining 45 respondents were not included because the pre-planned quota of completed participants had been reached by the time these individuals had completed the screening stage. All 400 participants (120 with type 1 diabetes mellitus and 280 with type 2 diabetes mellitus) completed all survey questions.

Demographics and participant characteristics are presented in Table 1. The mean ± SD age of the total study sample was 54.5 ± 12.2 years, mean BMI was 29.9 ± 7.2 kg/m2 and 61.3% were men. Half of participants (50.5%) reported having an above-median income. Over half of the participants (53.0%) reported that their most recent HbA1c value was >7.0% (>53 mmol/mol).

3.1. Mealtime insulin usage and wastage behaviour

ealtime insulin usage characteristics for the total sample, and for subgroups based on ether or not participants reported wasting insulin, are summarised in Table 2. This section focuses on the results for the total sample, and the described data are mean ± SD values, unless indicated otherwise. In the total study sample, participants had been receiving mealtime insulin for 8.6 ± 7.8 years. Just over half (53.0%) used a disposable prefilled pen, while 47.0% used a reusable pen with disposable cartridge. In response to questions about their use of insulin with meals, participants stated they took 54.8 ±

34.1 units of mealtime insulin per day, which was administered using 3.7 ± 5.9 injections per day (mean ± SE 3.7 ± 0.3 (95% CI 3.1-4.3) in bootstrapped analysis). Each month, participants used 11.3 ± 18.0 pens/cartridges.

When insufficient insulin remained in a prefilled pen/cartridge to provide a full dose, 63.5% of the total study sample reported that they had split the dose across two prefilled pens/cartridges (i.e. administered two injections); this subgroup of participants indicated that this occurred 2.1 ± 1.7 times per month. In addition, 15.0% of sample reported that they injected only what remained in their current prefilled pen/cartridge (i.e. a smaller-than-prescribed dose), and this subgroup indicated this happened 1.4 ± 1.3 times per month (Table 2).

Overall, amongst the total study sample, 145 participants (36.3%) indicated that they had discarded at least one pen/cartridge still containing insulin (i.e. wasted insulin), and 255 participants (63.7%) reported that they did not waste any insulin. The number of pens discarded still containing insulin among the total study sample was 2.0 ± 5.6 pens/month per participant. Among the subgroup who reported wasting insulin, 5.5 ± 8.2 pens containing unused insulin were discarded per participant each month. In each discarded pen, 8.6 ± 8.7 units remained, resulting in a mean ± SE of 51.0 ± 8.2 (95% CI 34.8-67.1) units of wasted insulin per participant per month in the subgroup who reported wasting some insulin (bootstrapped analysis).

When looking at participants' behaviour within individual countries, the proportion of the total country-specific sample who reported that they had administered two separate injections to complete a full dose (by splitting the dose across two prefilled pens/cartridges) was 82.0% in Germany, 69.0% in the UK, 53.0% in France and 50.0% in Italy, while the percentage who had omitted part of their prescribed dose by using only what remained in their current prefilled pen/cartridge was 13.0% in Germany, 16.0% in the UK, 12.0% in France and 19.0% in Italy. The mean ± SE total number of insulin units discarded per participant per month based on the total country-specific

study sample was 67.7 ± 24.5 (95% CI 19.7-115.7) in the UK, 49.6 ± 13.5 (95% CI 23.2-76.1) in France, 47.2 ± 11.4 (95% CI 25.0-69.5) in Italy and 24.7 ± 8.1 (95% CI 8.8-40.6) in Germany (bootstrapped analysis).

3.2. Comparison of participant characteristics and outcomes between those who did and did not report wasting insulin

Demographics and characteristics of participants who did and did not report wasting insulin are shown in Table 1. Individuals who reported wasting insulin were on average younger, had a lower BMI and were more often female than those who reported no wastage. In addition, they were more likely to be in employment and have an above-median household income. Participants who wasted insulin were less likely to be receiving oral antidiabetes medication or non-insulin injectable medication.

Insulin usage characteristics are presented in Table 2 according to whether or not participants reported wasting insulin. Participants who reported wasting insulin used significantly more prefilled pens/cartridges per month and incurred significantly greater out-of-pocket costs associated with insulin use. When insufficient insulin remained in a prefilled pen/cartridge to provide a full dose, significantly fewer participants who reported wasting insulin used two prefilled pens/cartridges to administer a full dose. However, there was no difference between the groups with respect to the proportion of participants who reported just taking what remained in their current prefilled pen/cartridge (Table 2).

HRQoL was similar between the two groups, as indicated by mean ± SD EQ-5D-5L index scores of 0.7 ± 0.3 for participants who reported wasting insulin and 0.8 ± 0.3 for those who reported wasting no insulin (p = 0.229). There were no significant differences between the two groups on any of the five EQ-5D dimensions (mobility, self-care, usual activities, pain/discomfort and anxiety/depression).

Participants' responses to the PITQ (overall attitudes towards insulin use) are presented in Table 3 according to whether or not they reported wasting insulin. Mean PITQ and SEITQ scores were significantly lower for participants who reported wasting insulin than for those who did not, indicating that participants who wasted insulin had a

more negative overall perception about insulin therapy and were less confident about using insulin therapy.

Participants' attitudes towards their mealtime insulin use, according to whether or not they reported wasting insulin, are presented in Fig. 2. There were significant differences between the two groups in attitudes towards cost, injection issues, convenience, confusion and efficacy. Participants who reported wasting insulin considered it more frustrating, time-consuming and painful to inject twice and preferred to discard insulin rather than do so. In addition, they were more likely to discard a prefilled pen/cartridge due to feeling confused about whether or not it contained enough insulin for a full dose, and due to concerns that the insulin may have exceeded its expiration date. Participants who reported wasting no insulin were more concerned about not wasting medicine or throwing away money, and were more willing to inject themselves twice if necessary to avoid waste.

Results of the comparison between people with type 1 diabetes mellitus and those with type 2 diabetes mellitus (with respect to wastage and other characteristics) are provided as online supplementary material.

[12]- 1

4. Discussion

This study sought to fill a gap in the medical literature about the extent of mealtime insulin wastage and the behaviour of people with type 1 and type 2 diabetes mellitus when insufficient insulin remains in a prefilled pen/cartridge to administer a full dose of mealtime insulin. It was found that transitions from used to new prefilled pens/cartridges often resulted in participants administering additional inj smaller-than-recommended doses, or discarding prefilled pens/cartridges that still contained insulin.

Very few data about how people with diabetes who self-administer insulin behave around pen transitions are available in the medical literature. In a single clinic study conducted 20 years ago, people with type 1 or 2 diabetes mellitus who were administering premixed insulin (30% short-acting/70% intermediate-acting) via a prefilled pen or cartridge pen were asked what they did when the reservoir was approaching exhaustion [12]. More than two-thirds of participants did not take any action to avoid wasting insulin. Approximately 5% of the 88 participants using prefilled pens (generally containing 300 units of insulin) wasted more than 10% of their prescribed insulin, and this figure rose to 36% amongst the 22 participants who used reusable cartridge pens (generally containing 150 units of insulin). Overall insulin wastage was greater with the smaller volume cartridge pens (estimated at a mean of units/participant/year) than with the prefilled pens (831 units/participant/year) . The current study contributes to the literature on this topic, and provides data on people who were specifically using mealtime insulin.

When people with diabetes who are taking mealtime insulin using a prefilled pen/cartridge come towards the end of a pen/cartridge, they must decide what action to take if there is not enough insulin remaining in their current pen/cartridge to provide a full dose. Some instructions for people with diabetes suggest that they either give a

prefil the m

partial dose using their current pen and the remainder of the dose using a new pen or else give the full dose with a new pen [8]. The former course of action means that no insulin is wasted, but the individual must administer two injections. Some people with diabetes may be willing to administer extra insulin injections in this circumstance, but many find self-injecting medication to be uncomfortable and inconvenient [23], and increasing the number of injections can have an adverse effect on HRQoL among people with diabetes [10]. Thus, improving the injection experience, through reducing the need for additional injections, could be of great benefit to people with diabetes.

In this survey, two-thirds of the participants reported that they had used two separate injections to provide a full dose (i.e. no insulin wasted). These results contrast with the study by Steel et al. [12], in which approximately 25% of participants reported regularly administering two separate injections of premixed insulin to obtain the full dose. Reasons for this disparity could include that premixed insulin is administered differently from mealtime insulin with respect to the number of daily injections as well as the number of units per injection. In light of this, and other potential issues such as the smaller sample size and the increase in the cost of insulin since 1997, it is not appropriate to extrapolate the findings of Steele et al. [12] to users of mealtime insulin.

The second option, giving the full dose with a new pen, means that only one injection is administered, but any insulin remaining in the current pen will be wasted. One-third of the survey respondents in this study reported that they recalled having discarded

led pens/cartridges that still contained insulin in a typical month. We have reported mean number of pens/units wasted for the overall sample as well as for the subset who actually reported wasting some insulin, because it may be helpful for decision-makers/payers to know the average number of prescribed units/pens being wasted by the total population receiving mealtime insulin as well as by individual patients. A third option that some people might choose when reaching the end of a prefilled pen/cartridge is to administer only what remains in the current pen (i.e. to take less

than the recommended dose). In this survey, 15.0% of participants reported that, when they reached the end of a prefilled pen/cartridge, they recalled having only used the insulin remaining in their current pen even if it meant injecting a smaller-than-recommended dose. In the study by Steel et al. [12], approximately 5% of participants reported taking less than their usual dose when they reached the end of a pen/cartridge. This is not a desirable option, given that poor adherence to antidiabetic treatment can be associated with glycaemic variability and worse glycaemic control [9,11,24]. Health professionals involved in the management of people with diabetes may need to take steps to individualise education to address the different pen transition behaviours, including the impact of reducing one's recommended dose. Understanding how often they omit the dose and the size of the omitted dose is relevant with respect to the impact on glycaemic control and thus will help determine how to individualise counselling for the appropriate patient.

II^dJI

four Europ<

The survey included participants from four European countries to provide a broader picture of insulin pen use practice. Statistical comparisons between the different countries were not performed, but the mean monthly quantity of insulin wasted in association with the transition from a used to a new prefilled pen/cartridge was lower in Germany than in other countries. It is possible that this could relate to the high proportion of participants in Germany who reported splitting a dose across two prefilled pens/cartridges, as this behaviour would limit the amount of wastage. The reasons for the difference were not investigated. Behaviour could be affected by factors that differ between countries, such as healthcare payment systems, out-of-pocket costs, cultural attitudes and national policies towards the environment and waste [6,25,26,27]. For example, differences in the level of coverage available in countries with healthcare insurance or national healthcare systems with prescription plans [6,25] will have cost implications for people with diabetes and could affect their decision about whether to use all of the insulin in a prefilled pen or not.

The study identified a number of characteristics, perceptions and attitudes that appeared to be associated with insulin wastage behaviour when insufficient insulin remained in a prefilled pen/cartridge to administer a full dose. Women, younger people and those with a lower BMI were more likely to report wasting insulin. In addition, participants who reported wasting insulin were more likely to be in employment and have an above-median household income than those who did not waste any insulin, but there was no difference in educational level. The length of time since being diagnosed with diabetes did not affect the likelihood of wasting insulin (Table 1). Participants who wasted insulin were less likely to be seeing the "other" category of diabetes healthcare provider, a subgroup that potentially included diabetes educators. Little has been published previously about the attitudes of people with diabetes towards insulin wastage, so comparison with other studies is not possible. Attitudes to treatment may influence decisions about how to manage transitions between pens. Reports differ with respect to gender-related differences in adherence [28,29], but women may have greater fear of injections than men [30], which could suggest they might prefer to waste insulin rather than administer two injections. Moreover, it can be hypothesised that people who are financially better off may be less concerned about the potential costs associated with wasting insulin in a society where individuals bear the cost of their prescription medications.

ition me< jnd n

The study found no difference in HRQoL between the groups who did and did not report wasting insulin. However, participants who wasted insulin had more negative perceptions about insulin therapy and were less confident about using insulin therapy. Participants who wasted insulin indicated that they considered administering two injections to be frustrating, time-consuming and painful and preferred to discard insulin if necessary to avoid doing so. These results suggest potential opportunities for focused education measures for people with diabetes. The management of diabetes is often complex and requires a multifaceted approach, but self-management is an

essential component [31]. This requires good understanding of, and commitment to, the prescribed regimen on the part of the person with diabetes. Simple, clear treatment regimens that can be followed without the need for decisions about dose-splitting, dose reduction or wastage may be helpful, and potentially may reduce some of the treatment burden for individuals with diabetes.

n 20 units practice

The study enrolled people with diabetes who were administering more than 20 mealtime insulin per day. Insulin doses of this level are relevant in clinical pr [32,33,34,35,36] and provide a reasonable chance of detecting and quantifying any wastage; the level of wastage would be harder to quantify for people who were taking only a small total amount of insulin per day. In addition, understanding the insulin wastage habits of people with higher dosing needs is particularly relevant because they may be candidates for new higher-strength formulations of mealtime insulin. Existing prefilled pens and cartridges contain 100 units/ml insulin, resulting in a maximum total of 300 units of insulin per prefilled pen/cartridge. A new bioequivalent 200 unit/ml formulation of rapid-acting insulin has been developed [14], which provides double the amount of insulin, 600 units/pen, in the same volume as a 3.0 ml-sized prefilled pen. The greater the number of insulin units contained in a single pen/cartridge, the less frequently the person will need to transition from an empty to a full pen. Steel et al. [12] found that much less premixed insulin was wasted when using a 300-unit (3.0 ml) disposable pen (3.6%) than when using a 150-unit (1.5 ml) cartridge (13.6%). Insulin pens that could offer a greater number of units per pen, whether by increasing the volume of the pen or by using higher-strength insulin formulations, could potentially reduce the number of times a person will be faced with too little insulin to complete their injection, and thus result in less total insulin wastage. This option may be particularly appropriate for people who could benefit from simplifying their insulin regimen by changing pens less often or those requiring a large number of units of mealtime insulin. Focused education measures (perhaps provided by pharmacists or

diabetes educators) about transitions between pens may also benefit people using insulin.

>hics of the ecruited, si

instri

avail ■

The study used a web-based survey. This type of online survey has been used successfully for other studies involving people with type 1 or type 2 diabetes mellitus [37,38,39,40,41]. Nonetheless, the study has several limitations. As for all cross sectional surveys, causality cannot be established. Although the demographics of the participants were as expected, a random sample of insulin users was not recruited, so the results may not be generalisable. In addition, older age groups tend to be under-represented in online research panels, which could have skewed the type 2 diabetes mellitus subgroup towards a younger-than-usual age sample. The average age of people with type 2 diabetes mellitus in the survey was 56.3 years (Table 1). Panels tend to have a higher representation of women and people in the middle socioeconomic range compared with the general population. However, there were more men than women in the current study sample, which is consistent with the reported higher prevalence of diabetes among adult men compared with women in European populations [42]. As with all online survey studies, responses were not confirmed by an interviewer. Clinical confirmation of data was not obtained. Although the survey was pilot tested to maximise comprehension, it is not possible to be sure how every participant interpreted each item. This could potentially explain some outlier values, such as seen for the average number of pens used per month. A validated nstrument to assess mealtime insulin usage and wastage behaviours was not

ilable, and therefore a newly developed set of questions was used for this section of the survey. However, pilot interviews were performed to ensure these new items could be understood by participants and were comprehensive in their evaluation of usage/wastage behaviours. The survey did not consider all possible options open to patients when transitioning between pens, but rather identified and examined the most common behaviours, which also reduced survey burden by ensuring the total survey

the fu frustri

length was acceptable to participants. Finally, it is possible that diabetes type could be a confounding factor with respect to the differences seen between subgroups of people who did and did not report wasting insulin.

Possible areas of future research include confirmation of the results in a more robust study, and additional work to fully understand the reasons underlying behavioural choices around pen transitions, and to examine whether educational approaches could modify such behaviour. The potential effect of higher-strength insulin formulations, allowing a greater number of units of insulin per pen, on transitions between pens could be evaluated. Differences in wastage behaviour between countries could be examined in more depth, as could the impact of pen transition behaviours on costs and HRQoL. Additionally, it would be interesting to further evaluate the behaviours of patients who injected only what remained in their current prefilled pen/cartridge, to determine what proportion of the prescribed dose they actually administered, whether they adjusted their eating to compensate for the lower insulin dose, how often this behaviour occurred, and whether it had any impact on their overall glycaemic control.

In conclusion, this study shows that people with type 1 or type 2 diabetes taking more than 20 units/day of mealtime insulin with a 100 units/ml pen/cartridge can find transitions between pens challenging. Two-thirds of participants administered two injections to obtain a full dose, 15.0% administered less than the prescribed dose, and one-third discarded a prefilled pen/cartridge that still contained insulin in order to take dose with a new prefilled pen/cartridge. Participants who wasted insulin found it ustrating, time-consuming and painful to inject themselves twice. These results highlight the need to identify ways of improving transitions between insulin pens so as to make transitions easier for insulin users, which has the potential to improve adherence to prescribed insulin doses and reduce waste. Possible options include developing educational strategies that address pen transition behaviour, individualising

strategies for managing transitions, and considering the use of alternative insulin formulations.

Acknowledgments

The study was funded by Eli Lilly. The authors thank Katherine Croom and Sue Williamson (Rx Communications, Mold, UK) for medical writing assistance during the preparation of this manuscript, which was funded by Eli Lilly.

Contributions

KVB has made substantial contributions to the conception and design of the work, acquisition, analysis and interpretation of the data, critical revision of the manuscript for important intellectual content, obtaining funding and providing administrative, technical and material support

RP has made substantial contributions to the acquisition, analysis and interpretation of the data for the work, critical revision of the manuscript for important intellectual content, assisting with statistical analysis and supervising the development of the manuscript

made subst evision of the

rative, techni< —p

JR and SMC have made substantial contributions to the conception and design of the work, critical revision of the manuscript for important intellectual content and providing administrative, technical and material support

All authors have participated sufficiently in the work to take public responsibility for the manuscript, give final approval for the manuscript to be published and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Declaration of interests

Kate Van Brunt and Sheila M Corrigan are employees and shareholders of Eli Lilly and Company. Riccardo Pedersini and Jillian Rooney have no relevant declarations.

References

[1] American Diabetes Association. 7. Approaches to glycemic treatment. Diabetes Care 2015;38 Suppl 1:S41-8.

[2] Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E, Nauck M, et al.

Management of hyperglycemia in type 2 diabetes, 2015: a patient-centred approach. Update to a position statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care 2015;38:140-9.

[3] Ng M, Fleming T, Robinson M, Thomson B, Graetz N, Margono C, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 19802013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 2014;384(9945):766-81.

e Europea

Global, re ' ' nd adults during tudy 201

[4] Watson L, Wilson BP, Alsop J, Kumar S. Weight and glycaemic control in type 2 diabetes:

what is the outcome of insulin initiation? Diabetes Obes Metab 2011 ;13(9):823—31.

[5] Lamos EM, Younk LM, Davis SN. Concentrated insulins: the new basal insulins. Ther Clin

Risk Manag. 2016;12:389-400.

[6] Perfetti R. Reusable and disposable insulin pens for the treatment of diabetes:

understanding the global differences in user preference and an evaluation of inpatient insulin pen use. Diabetes Technol Ther 2010;12 Suppl 1 :S79-85.

[7] Perez-Nieves M, Jiang D, Eby E. Incidence, prevalence, and trend analysis of the use of

insulin delivery systems in the United States (2005 to 2011). Curr Med Res Opin 2015;31 (5):891-9.

[8] Eli Lilly and Company. KwikPen™ insulin delivery device: user manual, 2013 September. Eli

Lilly and Company, Basingstoke, UK.

[9] Donnelly LA, Morris AD, Evans JM; DARTS/MEMO collaboration. Adherence to insulin and

Cits association with glycaemic control in patients with type 2 diabetes. QJM 2007;100(6):345-50.

[10] Evans M, Jensen HH, Bogelund M, Gundgaard J, Chubb B, Khunti K. Flexible insulin dosing improves health-related quality-of-life (HRQoL): a time trade-off survey. J Med Econ 2013;16(11):1357-65.

[11] Randlov J, Poulsen JU. How much do forgotten insulin injections matter to hemoglobin a1c

in people with diabetes? A simulation study. J Diabetes Sci Technol 2008;2(2):229-35.

[12] Steel JM, Carmichael C, Duncan C. Insulin wastage using a fixed mix of insulin with a pen:

the practice of patients in one clinic. Pract Diab Int 1997;14(6):157-8.

[13] Rees TM, Lennartz AH, Ignaut DA. A comparison of glide force characteristics between 2

prefilled insulin lispro pens. J Diabetes Sci Technol 2015;9(2):316—9.

[14] De la Peña A, Seger M, Soon D, Scott AJ, Reddy SR, Dobbins MA, et al. Bioequivalence

and comparative pharmacodynamics of insulin lispro 200 units/ml relative to insulin lispro (Humalog®) 100 units/mL. Clin Pharm Drug Devel 2016;5(1):69-75.

[15] Kantar Health. National health and wellness survey, 2015 (EU). Princeton, NJ: Kantar Health; 2015.

[16] Janssen MF, Pickard AS, Golicki D, Gudex C, Niewada M, Scalone L, et al. Measurement

properties of the EQ-5D-5L compared to the EQ-5D-3L across eight patient groups: a multi-country study. Qual Life Res 2013;22(7):1717-27.

[17] Hayes RP, Fitzgerald JT. Perceptions and attitudes are primary contributors to insulin

delivery system satisfaction in people with type 2 diabetes. Diabetes Technol Ther 2009;11 (7):419-26.

[18] Hayes RP, Naegeli AN. The contribution of pretreatment expectations and expectation-

perception difference to change in treatment satisfaction and end point treatment satisfaction in the context of initiation of inhaled insulin therapy in patients with type 2 diabetes. Diabetes Technol Ther 2010;12(6):447-53.

[19] Naegeli AN, Hayes RP. Expectations about and experiences with insulin therapy contribute

to diabetes treatment satisfaction in insulin-naive patients with type 2 diabetes. Int J Clin Pract 2010;64(7):908-16.

[20] Hayes RP, Curtis B, Ilag L, Nelson DR, Wong M, Funnell M. 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):358-67.

[21] International Diabetes Federation. IDF Diabetes Atlas, 7th edition. International Diabetes

Federation, 2015.

[22] Tanis

[23] Brod l\

EA, Hogg RV, Zimmerman DL. Probability and statistical inference. Pearson, 2015.

M, Kongso JH, Lessard S, Christensen TL. Psychological insulin resistance: patient beliefs and implications for diabetes management. Qual Life Res 2009;18(1 ):23-32.

[24] Murata GH, Duckworth WC, Shah JH, Wendel CS, Hoffman RM. Sources of glucose

variability in insulin-treated type 2 diabetes: the Diabetes Outcomes in Veterans Study (DOVES). Clin Endocrinol (Oxf). 2004 Apr;60(4):451-6.

[25] Magnolti MA, Rayfield EJ. An update on insulin injection devices. Insulin 2007;2(4):173-81

[26] Tong AY, Peake BM, Braund R. Disposal practices for unused medications around the

world. Environ Int 2011 ;37(1):292-8.

[27] Bound JP, Kitsou K, Voulvoulis N. Household disposal of pharmaceuticals and perception

of risk to the environment. Environ Toxicol Pharmacol 2006;21 (3):301-7.

[28] Davies MJ, Gagliardino JJ, Gray LJ, Khunti K, Mohan V, Hughes R. Real-world factors affecting adherence to insulin therapy in patients with Type 1 or Type 2 diabetes mellit

a systematic review. Diabet Med 2013;30(5):512-24.

[29] Peyrot M, Barnett AH, Meneghini LF, Schumm-Draeger PM. Factors associated wi

injection omission/non-adherence in the Global Attitudes of Patients and Physician s in Insulin Therapy study. Diabetes Obes Metab 2012;14(12):1081-7.

[30] Nam S, Chesla C, Stotts NA, Kroon L, Janson SL. Factors associated with psychological

insulin resistance in individuals with type 2 diabetes. Diabetes Care 2010;33(8):1747-9.

[31] Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E, Nauck M, et al.; American

Diabetes Association (ADA); European Association for the Study of Diabetes (EASD). Management of hyperglycemia in type 2 diabetes: a patient-centered approach: position statement of the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care 2012;35(6):1364-79.

[32] Davidson MB, Raskin P, Tanenberg RJ, Vlajnic A, Hollander P. A stepwise approach to

insulin therapy in patients with type 2 diabetes mellitus and basal insulin treatment failure. Endocr Pract 2011 ;17(3):395-403.

[33] Bebakar WM, Lim-Abrahan MA, Jain AB, Seah D, Soewondo P. Safety and effectiveness of

insulin aspart in type 2 diabetic patients: results from the ASEAN cohort of the Achieve study. Diabetes Res Clin Pract 2013;100 Suppl 1:S17-23.

[34] Aagren M, Luo W, Moës E. 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.

Holleman F, Schmitt H, Rottiers R, Rees A, Symanowski S, Anderson JH. Reduced frequency of severe hypoglycemia and coma in well-controlled IDDM patients treated with insulin lispro. The Benelux-UK Insulin Lispro Study Group. Diabetes Care 1997;20(12):1827-32.

[36] Garg SK, Rosenstock J, Ways K. Optimized basal-bolus insulin regimens in type 1

diabetes: insulin glulisine versus regular human insulin in combination with Basal insulin glargine. Endocr Pract 2005;11 (1 ):11—7.

[37] Mitchell BD, Viet ri J, Zagar A, Curtis B, Reaney M. Hypoglycaemic events in patients with

type 2 diabetes in the United Kingdom: associations with patient-reported outcomes and self-reported HbAlc. BMC Endocr Disord 2013;13:59.

[38] Lopez JM, Annunziata K, Bailey RA, Rupnow MF, Morisky DE. Impact of hypoglycemia on

patients with type 2 diabetes mellitus and their quality of life, work productivity, and medication adherence. Patient Prefer Adherence 2014;8:683-92.

[39] Lopez JM, Bailey RA, Rupnow MF, Annunziata K. Characterization of type 2 diabetes mellitus burden by age and ethnic groups based on a nationwide survey. Clin Tl 2014;36(4):494-506.

. Clin Ther

[40] Reach G, Le Pautremat V, Gupta S. Determinants and consequences of insulin initiation for

type 2 diabetes in France: analysis of the National Health and Wellness Survey. Patient Prefer Adherence 2013;7:1007-23.

[41] Marrett E, Radican L, Davies MJ, Zhang Q. Assessment of severity and frequency of self-

reported hypoglycemia on quality of life in patients with type 2 diabetes treated with oral antihyperglycemic agents: A survey study. BMC Res Notes 2011 ;4:251.

[42] NCD Risk Factor Collaboration (NCD-RisC). Worldwide trends in diabetes since 1980: a

pooled analysis of 751 population-based studies with 4.4 million participants. Lancet. 2016 Apr 9;387(10027):1513-30.

Figure 1 - Participant disposition. [[See separate file]]

Figure 2 - [[See separate file This figure should be reproduced in colour.]] Attitudes towards mealtime insulin use by wastage behaviour. Based on newly developed questions assessing mealtime insulin use (see online supplementary material). Ratings go from 1 'strongly disagree' (darker red) to 4 'neither agree nor disagree' (grey) to 7 'strongly agree' (darker blue). The average agreement/disagreement within the 'waster' and 'non-w; subgroups is reported within the circles: values range from -100% (complete disagreement, corresponding to a rating of 1) via 0% (neutral, corresponding to a rating of 4) to 100% (complete agreement, corresponding to a rating of 7). * p < 0.05 (independent-sample t test).

ependent

Table 1 - Participant demographics and characteristics according to type of diabetes and according to whether or not participants reported wasting insulin.

Parameter

T1DM T2DM Total Participants reporting

(N = 120) (N = 280) (N = 400) NO Wastage

wastage (n = 145) (N = 255)

Age, years

50.3 ± 12.9*

56.3 ± 11.4* 54.5 ± 12.2 56.5 ± 11.7

50.9 ± 12.2

18-44 43 (35.8)* 45 (16.1)* 88 (22.0) 37 (14.5) 51 (35.2) 0.000

45-64 57 (47.5)* 166 (59.3)* 223 (55.8) 150 (58.8) 73 (50.3)

>65 20 (16.7)* 69 (24.6)* 89 (22.3) 68 (26.7) 21 (14.5)

Gender

Male 80 (66.7) 165 (58.9) 245 (61.3) 169 (66.3) 76 (52.4) 0.006

Female 40 (33.3) 115 (41.1) 155 (38.8) 86 (33.7) 69 (47.6)

BMI, kg/m2 26.9 ± 4.9* 31.3 ± 7.7* 29.9 ± 7.2 30.6 ± 7.8 28.7 ± 5.8 0.014

Employed (full- or part-time), yes 65 (54.2)* 116 (41.4)* 181 (45.3) 104 (40.8) 77 (53.1) 0.017

Completed university, yes 51 (42.5) 92 (32.9) 143 (35.8) 83 (32.5) 60 (41.4) 0.076

Household income 0.044

Below median 37 (30.8) 113 (40.4) 150 (37.5) 106 (41.6) 44 (30.3)

Above median 70 (58.3) 132 (47.1) 202 (50.5) 117 (45.9) 85 (58.6)

Declined to answer 13 (10.8) 35 (12.5) 48 (12.0) 32 (12.5) 16 (11.0)

Years since diagnosis 23.4 ± 14.2* 13.0 ± 8.8* 16.1 ± 11.7 16.8 ± 11.2 14.9 ± 12.4 0.123

Non-insulin diabetes treatmentb V

Diet and exercise 48 (40.0) 129 (46.1) 177 (44.3) 112 (43.9) 65 (44.8) 0.861

Oral antidiabetic medication 13 (10.8)*

Non-insulin injectable medication 2 (1.7)*

Other 3 (2.5)* Last HbA1c

143 (51.1)* 26 (9.3)* 0*

156 (39.0) 28 (7.0) 3 (0.8)

109 (42.7) 23 (9.0) 2 (0.8)

47 (32.4) 0.042 5 (3.4) 0.036 1 (0.7)

<6.5-7.0% (<48-53 mmol/mol) 55 (45.8) 75 (26.8) 130 (32.5) 88 (34.5) 42 (29.0)

7.1-9.5% (54-80 mmol/mol) 43 (35.8) 151 (53.9) 194 (48.5) 120 (47.1) 74 (51.0)

9.6 to >10.6% (81 to >92 mmol/mol) 3 (2.5) 15 (5.3) 18 (4.5) 10 (3.9) 8 ( 5.5)

Not known 19 (15.8) 39 (13.9) 58 (14.5) 37 (14.5) 21 (14.5)

Exercised in past monthc, yes 81 (67.5) 171 (61.1) 252 (63.0) 162 (63.5) 90 (62.1) 0.771

Days exercised per month 7.9 ± 8.4 7.2 ± 8.8 7.4 ± 8.7 7.8 ± 9.1 6.6 ± 7.9 0.155

Taking steps to lose weight, yes 36 (30.0)* 130 (46.4)* 166 (41.5) 106 (41.6) 60 (41.4) 0.971

Diabetes healthcare providerb

Primary care physician 77 (64.2) 198 (70.7) 275 (68.8) 175 (68.6) 100 (69.0) 0.944

Hospital-based diabetes specialist 84 (70.0)* 161 (57.5)* 245 (61.3) 149 (58.4) 96 (66.2) 0.125

Practice nurse 13 (10.8) 38 (13.6) 51 (12.8) 34 (13.3) 17 (11.7) 0.643

Diabetic nurse specialist 22 (18.3) 53 (18.9) 75 (18.8) 48 (18.8) 27 (18.6) 0.963

Dietitian 14 (11.7) 38 (13.6) 52 (13.0) 33 (12.9) 19 (13.1) 0.960

Other 17 (14.2) 37 (13.2) 54 (13.5) 44 (17.3) 10 (6.9) 0.004

CCI 2.3 (1.6)* 2.8 (1.8)* 2.6 (1.8) 2.7 ± 1.8 2.5 ± 1.7 0.454

Data are presented as mean ± SD or n (%).

BMI = body mass index, CCI = Charlson Comorbidity Index, HbA1c = glycated haemoglobin, SD = standard deviation, T1 DM = type 1 diabetes mellitus, T2DM = type 2 diabetes mellitus.

a P value for participants reporting insulin wastage versus participants reporting no wastage. b Choices were not mutually exclusive. c Exercised for more than 20 minutes per time.

* Significant difference between participants with type 1 and type 2 diabetes (p=0.019 to p=0.000).

Data are results from the survey items developed specifically for this study (see online supplementary material).

Table 2 - Insulin usage characteristics for prefilled pens/cartridges used for mealtime insulin according to type of diabetes and according to whether or not participants reported wasting insulin.

Parameter

T1DM (N = 120)

T2DM (N = 280)

Total (N = 400)

Participants reporting

No wastage (N = 255)

Wastage (N = 145)

General insulin usage Time on mealtime insulin, years

Type of pen

Cartridges

Prefilled pens

Number of prefilled pens/cartridgesb typically kept in total across all locations (home, work, etc)c

Number of injections typically taken per dayc

Number of units typically taken with meals per dayc

13.3 ± 9.4* 6.6 ± 5.9*

8.6 ± 7.8 [0.3-54.5]

58 (48.3) 130 (46.4) 188 (47.0) 125 (49.0) 62 (51.7) 150 (53.6) 212 (53.0) 130 (51.0)

8.9 ± 7.6

8.4 ± 5.0* 11.1 ± 11.5*

0.1 10.8 ± 10.6 90.0]

3.8 ± 4.9

......

3.6 ± 6.3 3.7 ± 5.9 [1.0-78.0]

3.7 ± 5.8

8.1 ± 8.1 0.305

63 (43.4) 82 (56.6) 9.6 ± 9.1

3.7 ± 6.0 0.905

60.7 ± 36.7* 54.8 ± 34.1 55.4 ± 33.5 53.7 ± 35.2 0.643

[21.0225.0]

Breakfast

Dinner

Number of prefilled pens/cartridgesb typically used per monthc

Out-of-pocket costs associated with insulin use (€)

Transitions between prefilled pens/cartridgesb

Use two prefilled pens/cartridgesb to provide one full dose

Times per month

Use only insulin remaining in current prefilled pen/ cartridgeb (i.e. less than recommended dose)

Times per month

Number of prefilled pens/cartridgesb used entirely (no insulin remaining) per month

11.0 ± 8.1* 18.9 ± 14.7* 16.5 ± 13.6 16.7 ± 13.3 16.1 ± 14.1 0.667

[0.0-75.0]

13.9 ± 8.4* 18.4 ± 14.6* 17.1 ± 13.2 17.4 ± 13.1

[0.0-75.0]

1.8 (1.7) 15 (12.5)

1.2 (0.9) 4.2 ± 6.2*

16.6 ± 13.5 0.598

16.2 ± 10.4* 23.4 ± 14.6* 21.2 ± 13.9 21.3 ± 13.6 21.0 ± 14.5 0.830

[0.0-75.0]

5.2 ± 6.7* 13.8 ± 20.5* 11.3 ± 18.0 9.7 ± 14.8

[1.0-99.0]

4.9 ± 12.7 18.5 ± 63.4 14.2 ± 53.1 8.1 ± 21.6

......

[0.0-650.0]

76 (63.3) 178 (63.6) 254 (63.5) 196 (76.9)

2.2 (1.8 45 (16.1

(1.7) 60 (15.0)

1.5 (1.4) 1.4 (1.3)

11.4 ± 17.6* 9.3 ± 15.5 [0.0-99.0]

2.0 (1.7) 35 (13.7)

1.2 (1.2) 9.7 ± 14.8

1 ± 22.3 0.018

25.4 ± 84.0 0.034

58 (40.0) 0.000

2.3 (1.9) 25 (17.2)

1.6 (1.4) 8.6 ± 16.6

0.346 0.344

0.215 0.508

Data are presented as mean ± SD or n (%) for T1 DM, T2DM and participants reporting no wastage or wastage, and as mean ± SD [range] or n (%) for total participants. SD = standard deviation, T1 DM = type 1 diabetes mellitus, T2DM = type 2 diabetes mellitus. a P value for participants reporting insulin wastage versus participants reporting no wastage.

ellitus, T ersus pa

b Disposable prefilled pen/reusable pen with disposable cartridge used for administration of mealtime insulin. c Questions refer to current usage.

* Significant difference between participants with type 1 and type 2 diabetes (p = 0.029 to p = 0.000). Data are results from the survey items developed specifically for this study (see supplementary material).

Table 3 - Attitudes and confidence regarding overall insulin usage, assessed using the PITQ (incorporating the SEITQ), according to whether or not participants reported wasting insulin.

Parameter

Participants reporting

No wastage (N = 255)

PITQ total score 1

PITQ statements about insulin: "Taking insulin..." Makes it easier to control my blood sugars Restricts my life

Causes me to have severe episodes of low bloo< Makes me feel better Causes me to gain an undesirable amount of weight PITQ statements about IDS: "My IDS is." Physically painful

Easy for me to use away from home Is not noticed by others when I use it

3.4 3.0 5.6 4.4

ithers wh

2.7 5.7 4.6

5.9 3.7

3.6 5.4 4.4

3.7 5.3 4.6

It is easy to get the insulin dose I need with my IDS Is convenient SEITQ total score a

SEITQ statements about confidence level: "I am confident that I will be able to..." b

Take my insulin correctly

Avoid severe hypoglycaemic (low blood sugar) episodes when taking insulin

Consistently avoid low blood sugars when taking insulin

Consistently avoid high blood sugars when taking insulin

Avoid diabetes complications by taking insulin

5.6 76.8

5.8 5.3 71.9

6.2 5.4

4.9 5.1 5.4

Data are presented as means.

IDS = insulin delivery system, PITQ = Perceptions of Insulin Therapy Questionnaire, SEITQ = Self-efficacy about Insulin Therapy Questionnaire (PITQ section assessing confidence).

a Possible total PITQ or SEITQ score of 1-100. Lower score = more negative perception/lower confidence, higher score = more positive perception/higher confidence. b Possible score for each item of 1-7. * p = 0.000, t p = 0.008 for between-group comparison.

Highlights

Behaviours vary when insufficient insulin remains in a pen to provide a full dose. 63.5% of people with diabetes give two injections to complete a dose. 15.0% use only what remains in their current pen (less than prescribed). 36.3% discard pens containing unused insulin.

The latter group considers it frustrating, time-consuming and painful to inject twice.