If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author at: RWE Patient-Centered Endpoints Center of Excellence, IQVIA, 17 bis place des Reflets, Tour D2, TSA 64567, 92099 La Defense Cedex, France.
This study was conducted to evaluate existing burden and unmet needs related to mealtime insulin (MTI) injection timing among adult Japanese patients with type 1 (T1D) and type 2 (T2D) diabetes. It also aimed to evaluate if a novel MTI could reduce this burden.
Methods
This study comprised of a qualitative pilot study facilitating development of an online survey; followed by an online quantitative survey involving T1D, young T2D (yT2D) and elderly T2D (eT2D) patients to assess burden of current MTI timings in Japan.
Results
Overall, 38% patients (amongst T1D, yT2D and eT2D groups) reported injecting MTI just before start or during meal in the past month. Experiencing lower glucose level/hypoglycemic condition before the meal and forgetting were the main reasons for injecting during/after meal in T1D and T2D patients respectively. Patients reported moderate-to-severe burden in multiple aspects of their lives, associated with current MTI timing. Most patients perceived that this burden would remain the same if a faster acting MTI was available.
Conclusions
Substantial burden reported by Japanese patients regarding the current MTI timings suggests the need for new MTI products that could achieve optimal post-prandial glucose control at different timings to meet patients’ needs in Japan.
Diabetes mellitus (DM; hereafter referred to as diabetes) is a chronic endocrine disorder characterized by hyperglycemia, which occurs due to deficiency of insulin secretion from pancreas or insulin sensitivity in the body, or both [
]. Based on the type of defect in insulin physiology (secretion or sensitivity), diabetes can be clinically classified into two major types – type 1 diabetes (T1D) and type 2 diabetes (T2D) [
Duration of diabetes and types of diabetes therapy in Japanese patients with type 2 diabetes: The Japan diabetes complication and its prevention prospective study 3 (JDCP study 3).
]. Recombinant human insulin and insulin analogues (such as insulin glulisine and insulin lispro) approved as MTI products in Japan need to be administered up to 15 min prior to commencing a meal, according to their prescribing information [
] and ultra-rapid insulin products available in the US and Europe that can be administered immediately before a meal and up to 20 min after starting a meal [
], which have recently received marketing approval in Japan as well. Since the clinical effect of MTI products to control the post-prandial glucose level is optimally reached only if administered according to their prescribing information, administration of currently available MTI products during other timings, such as during or after a meal, are not recommended in Japan [
The challenges associated with the older MTI products available in Japan are related to their inflexible injection timings and individual patient-related factors. The latter includes factors such as lifestyle, forgetfulness, difficulty in administering insulin at the same time every day, in public places, during travel or change/maintenance of normal routine [
]. Many patients tend to administer MTI immediately before a meal, which is not the best timing to control post-prandial plasma glucose levels using existing MTIs (but may be appropriate for ultra-rapid insulin formulations). Although little evidence exists, it can be anticipated that some patients administer MTI after a meal [
The value of fast-acting insulin aspart compared with insulin aspart for patients with diabetes mellitus treated with bolus insulin from a UK health care system perspective.
Attitudes of patients and physicians to insulin therapy in Japan: an analysis of the global attitude of patients and physicians in insulin therapy study.
]. Inappropriate diet and dietary timing (including reduced food intake or skipping a meal) are some of the factors associated non-adherence to insulin injection before meals, which may lead to severe hypoglycemia in diabetic patients [
The current status of treatment-related severe hypoglycemia in Japanese patients with diabetes mellitus: A report from the committee on a survey of severe hypoglycemia in the Japan diabetes society.
The present study aimed to evaluate the existing burden and unmet needs related to injection timing of MTI among adult patients with diabetes in Japan who currently take MTI. In addition, the study aimed to evaluate what burdens may be reduced if an MTI could be administered 2 min prior to or within 20 min after the commencement of a meal.
2. Subjects, materials and Methods
2.1 Study design
The study was conducted in two phases among T1D and T2D patients in Japan: (1) a qualitative pilot study conducted on patients recruited from a private clinic (PS clinic) in Fukuoka, Japan; and (2) a quantitative online survey conducted on a panel of patients who were referred from an independent market research company (Medilead). The online survey questions were developed following a literature review, inputs from diabetologists and the qualitative pilot study. The selected patients in the online survey were categorized into three groups – T1D group, young T2D group (yT2D; comprising of patients <65 years of age) and elderly T2D group (eT2D; comprising of patients ≥65 years of age) to ensure representation of both diabetes types (T1D and T2D), as well as elderly patients who are less likely to participate in online surveys. This study only recruited patients receiving rapid insulin analogues.
2.2 Qualitative pilot study (for survey development)
The relevance and sufficiency of the questions were assessed through the qualitative pilot interview study, which comprised of one-to-one, in-person, semi-structured cognitive debriefing interviews lasting approximately 60 min involving 39 patients with diabetes. The interviews were conducted in Japanese, in two sets (N = 18, N = 21), to allow for modification and re-testing of the survey questionnaire after the first 18 patients. The interviews took place between July and September 2018 were conducted by trained interviewers in the PS clinic. Interviewers were trained in advance by the study team using the interview guide to ensure that the interviewers conducted each interview consistently, and that the appropriate results were reported in a uniform manner.
To be eligible to participate in the qualitative study, a patient should have been a resident of Japan, at least 20 years of age and should have been able to read and speak Japanese. To confirm the diagnosis, they must have been willing and able to bring their medication notebook to the interview or to describe how they were diagnosed with diabetes and how they are treated. They should also have been taking MTI regularly for at least three months. Patients using an insulin pump instead of multiple daily injections, having a diagnosis of gestational diabetes, directly involved with pharmaceutical companies or treating diabetes patients were not eligible for participation.
Two additional concepts were identified from the first set of interviews: (1) ‘blood glucose control’ which was included in final survey; and (2) ‘variation in food volume’, which was not included in the final survey to avoid redundancy with existing questions on food volume, timing, and planning. Additionally, one question concerning ‘anxiety’ was removed since most patients considered that this question was about hypoglycemia, which was considered redundant with the question on hypoglycemia already included in the questionnaire. No additional concepts were identified in the second set of interviews, only minor modifications were made to the language of other questions for clarity.
2.3 Online survey
The patients for the online survey were recruited from a patient panel in Japan between November and December 2018. Recruited patients were Japanese residents, at least 20 years of age, with a diagnosis of either T1D or T2D and taking MTI injection regularly for at least three months. Patients taking oral hypoglycemics or non-insulin injectables at the time of the survey were also allowed to participate in the study. Patients using an insulin pump, having a diagnosis of gestational diabetes and those directly involved with pharmaceutical companies or treating diabetes patients were excluded from the study. Diagnosis was confirmed by the upload of a photo from their medication notebook, but this requirement was removed during recruitment only for eT2D patients to encourage elderly participants (over 65 years old), who may have difficulty uploading the photo, to participate. The questions in the online survey included experience of hypoglycemia with MTI, lifestyle and the burden they felt related to the timings of the MTI. As the data were self-reported, patients were provided with the following definitions of hypoglycemia to facilitate their understanding of the terminologies: (1) severe hypoglycemia, defined as blood glucose level reaching less than 50 mg/mL with/without occurrence of hypoglycemic symptoms (such as unconsciousness), which may require assistance from another person, (2) non-severe hypoglycemia, defined as feeling hunger, sweating, shiver, fatigue and weakness due to low blood glucose.
No formal sample size calculation was performed as this was a descriptive study. Approximately 400 patients were considered appropriate for the online survey (approximately 100 T1D and 300 T2D patients). This was done to allow for reasonable description of concepts of burden and unmet needs related to injection timing of MTI and meal in Japan, and reasonable representation of subgroups of interest.
2.4 Statistical analysis
All analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA). Qualitative data from the cognitive debriefing interview stage were analyzed using Microsoft Excel for Office 365 (Microsoft Corp., Redmond, Washington, USA). Descriptive statistics (e.g., mean, standard deviation, frequencies) were used to summarize sociodemographic and clinical characteristics of the study sample, questions related to MTI experience and lifestyle, and burden of MTI timing. The relationship between questions capturing the MTI burden and how it may change with the introduction of a new faster MTI was explored using cross-tabulations and descriptive statistics. Multivariable logistic regression was used to explore the factors associated with MTI injection during or after meal in the past month.
2.5 Ethics statement
This study was conducted following a preapproved protocol submitted to the ethics committee. All analyses were performed following a predefined statistical analysis plan (SAP) prior to availability of study data. All these were in accordance with ethical guidelines from the Ministry of Education, Culture, Sports, Science and Technology, and the Ministry of Health, Labor and Welfare (MECSST and MHLW), Japan.
Ethical approval for this study was obtained from local ethics committees: Hakata Clinic Institutional Review Board for the qualitative pilot study and Specified Nonprofit Corporation Clinical Research Promotion Network, Japan for the online survey.
Patients who participated in the qualitative pilot study provided written informed consent prior to participation and those participating in the online survey gave consent electronically prior to responding to the questionnaire. Participation was voluntary, and patients were free to withdraw their consent at any time for any reason. Patients who participated in the qualitative pilot study or the online survey received remuneration for their time: in the pilot survey, the participants received 10,000 yen (approximately USD$92); and in the online survey, the participants received 3,000 yen (approximately USD$27).
3. Results
3.1 Patient characteristics
Overall, 382 patients (167 T1D patients, 143 yT2D patients and 72 eT2D patients) completed the online survey and their patient characteristics (by diabetes type) are presented in Table 1. Almost two-third of the T1D patients were males (mean age, 48 years). Over 83% (mean age, 51 years) and 87% (mean age, 70 years) of yT2D and eT2D patients respectively, were males. Almost three-quarters of the patients in T1D and yT2D groups were in day-time employment, whereas in the eT2D group, most patients were either in day-time employment, retired or unemployed. Further details are provided in Table 1 and Supplemental Table S 1.
Table 1Patient characteristics and demographics data.
Data presented for 166 patients. Key: ISI: Insulin self-injection; NISI: Non-insulin antidiabetic injectable; SD: Standard Deviation; T1D: Type 1 diabetes; T2D: Type 2 diabetes
, Mean (SD)
22.9 (3.6)
26.8 (8.2)
24.4 (3.8)
24.6 (6.0)
Employment status (categorized), n (%)
Day employment
129 (77.2)
104 (72.7)
26 (36.1)
259 (67.8)
Night-worker
2 (1.2)
0 (0.0)
1 (1.4)
3 (0.8)
Homemaker
14 (8.4)
7 (4.9)
5 (6.9)
26 (6.8)
Student
2 (1.2)
0 (0.0)
0 (0.0)
2 (0.5)
Retired
4 (2.4)
8 (5.6)
17 (23.6)
29 (7.6)
Unemployed
14 (8.4)
23 (16.1)
19 (26.4)
56 (14.7)
Other
2 (1.2)
1 (0.7)
4 (5.6)
7 (1.8)
Comorbidity, n (%)
Angina requiring hospitalization
5 (3.0)
8 (5.6)
7 (9.7)
20 (5.2)
Myocardial Infraction
5 (3.0)
13 (9.1)
8 (11.1)
26 (6.8)
Congestive heart failure
5 (3.0)
3 (2.1)
1 (1.4)
9 (2.4)
Stroke
1 (0.6)
8 (5.6)
5 (6.9)
14 (3.7)
Gastrectomy
2 (1.2)
2 (1.4)
2 (2.8)
6 (1.6)
None of the above
151 (90.4)
115 (80.4)
52 (72.2)
318 (83.2)
HbA1c level, n (%)
Below 7%
69 (41.3)
54 (37.8)
28 (38.9)
151 (39.5)
Between 7 and 7.9%
52 (31.1)
55 (38.5)
36 (50.0)
143 (37.4)
8% or higher
46 (27.5)
34 (23.8)
8 (11.1)
88 (23.0)
Time since first diagnosed with diabetes (years), Mean (SD)
15.2 (11.8)
12.6 (9.0)
17.5 (11.2)
14.7 (10.8)
Time since first treated with MTI via self-injection (years), Mean (SD)
10.7 (8.02)
7.7 (6.37)
8.8 (6.95)
9.2 (7.35)
Patients currently taking oral antidiabetics for treatment, n (%)
31 (18.6)
86 (60.1)
37 (51.4)
154 (40.3)
Approximate number of MTI injections in a regular day (categorized), n (%)
<3 injections per day
27 (16.2)
53 (37.1)
24 (33.3)
104 (27.2)
≥3 injections per day
140 (83.8)
90 (62.9)
48 (66.7)
278 (72.8)
* Data presented for 166 patients. Key: ISI: Insulin self-injection; NISI: Non-insulin antidiabetic injectable; SD: Standard Deviation; T1D: Type 1 diabetes; T2D: Type 2 diabetes
The mean time since diagnosis of diabetes for all patients was 14.7 years, while the mean time since first treatment with MTI via self-injection was 9.2 years. Oral hypoglycemics were consumed by about one-fifth of T1D patients, and over half of yT2D and eT2D patients. About a fifth of T1D and yT2D, and a tenth of eT2D patients had >8% HbA1c. Approximately three-quarters of patients measured their blood glucose levels daily.
A total of 1.6% presented with a history of gastrectomy while 3.1% of patients were being treated for cancer.
3.2 Hyperglycemia and hypoglycemia
Experience of hyperglycemic symptoms (such as fatigue, increased thirst, frequent urination and blurry vision) a few hours after meal in the past week were experienced by 75.7% of patients. The most commonly reported hyperglycemic symptoms reported by T1D, yT2D and eT2D patients were fatigue/tiredness and increased thirst. Other hyperglycemic symptoms reported were frequent urination and blurry vision (Supplemental Table S 1). More patients in the T1D group (29.3%) experienced at least one episode of severe hypoglycemia in the past year compared to patients in yT2D (12.6%) and eT2D (11.1%) groups. Similarly, non-severe hypoglycemia (three or more episodes in the last month) was reported more frequently by the T1D group (26.9%) than by the yT2D (4.2%) and eT2D (6.9%) groups. Nausea or vomiting in the previous month was reported in 11.8% of patients. The proportion of patients reporting hyperglycemic and hypoglycemic events are presented in Supplemental Table S 1.
3.3 MTI timing
Table 2 presents MTI timings reported by the patients. Patients were allowed to provide multiple responses to questions related to their normal MTI administration timing. In T1D and yT2D groups, approximately 40% patients reported normally injecting MTI between 3 and 15 min before their meal or at the start of meal. In the eT2D group, 61.1% of patients reported injecting MTI between 3 and 15 min before meal, whereas 27.8% reported injecting at start of meal. MTI was reportedly injected over 15 min before the start of meal by 15.0%, 12.6% and 12.5% of T1D, yT2D and eT2D patients respectively. MTI was normally injected after the start of meal by 13.2% and 6.3% and 4.2% of T1D, yT2D and eT2D patients respectively. Regarding timing adjustment, one-fifth of T1D and one-tenth of yT2D and eT2D patients reported that they normally change the injection timing based on their schedule, physical condition and blood glucose level.
Table 2Patient report of mealtime insulin injection timing.
Patients were asked “In the past month, how was the approximate percentage of mealtime insulin timing?” Patients responded by entering numbers between 0 and 100 for each timing to equal 100%. The mean describes the average percentage of the time that patients inject before, at start and after meal..Key: BG: Basal Glucose; HCP: Healthcare Provider; MTI: Mealtime Insulin; SD: Standard Deviation; T1D: Type 1 diabetes; T2D: Type 2 diabetes.
Over 15 mins before meal
13.8 (28.9)
18.7 (34.8)
15.1 (30.8)
15.9 (31.6)
Between 3 and 15 min before start of meal
31.2 (36.1)
33.7 (40.8)
55.4 (42.5)
36.7 (40.1)
At start of meal to 2 min before start of meal
42.8 (39.1)
35.3 (41.7)
23.9 (36.4)
36.4 (40.1)
After start of meal
12.2 (23.6)
12.4 (29.4)
5.6 (15.6)
11.0 (24.9)
* This was a multiple-choice question i.e., patients were able to provide multiple answers for the same question.
** Patients were asked “In the past month, how was the approximate percentage of mealtime insulin timing?” Patients responded by entering numbers between 0 and 100 for each timing to equal 100%. The mean describes the average percentage of the time that patients inject before, at start and after meal..Key: BG: Basal Glucose; HCP: Healthcare Provider; MTI: Mealtime Insulin; SD: Standard Deviation; T1D: Type 1 diabetes; T2D: Type 2 diabetes.
Overall, 37.7% of patients reported having dosed during or after meal in the past month. Over half of T1D patients (52.1%) and about a quarter of yT2D and eT2D patients (28.0% and 23.6 respectively) reported having dosed during or after meal in the past month (Table 3).
Table 3Patient report of adherence to mealtime insulin injection in past week.
T1D (N = 167)
Young T2D, <65yrs (N = 143)
Elderly T2D, ≥65yrs (N = 72)
Overall sample (N = 382)
Experience of dosing during or after meal in the past month, n (%)
87 (52.1)
40 (28.0)
17 (23.6)
144 (37.7)
If yes, Reason for experiencing dosing during or after meal in the past month, n (%)
Percentages do not add to 100 as patients may report more than one outcome. The denominators of the proportions are the number of patients with experience of dosing after a meal.
Forgot to administer before meal
19 (21.8)
17 (42.5)
11 (64.7)
47 (32.6)
Physician instruction
7 (8.0)
8 (20.0)
1 (5.9)
16 (11.1)
Ate more than planned
16 (18.4)
7 (17.5)
0 (0.0)
23 (16.0)
Appetite fluctuates/Not sure how much I would eat before a meal
15 (17.2)
3 (7.5)
0 (0.0)
18 (12.5)
Too busy to prepare before meal
14 (16.1)
6 (15.0)
4 (23.5)
24 (16.7)
Schedule of other oral medicines
0 (0.0)
1 (2.5)
0 (0.0)
1 (0.7)
Stayed outside / worked outside
19 (21.8)
6 (15.0)
5 (29.4)
30 (20.8)
Couldn’t find a place to inject
15 (17.2)
7 (17.5)
4 (23.5)
26 (18.1)
Ate outside of home
27 (31.0)
12 (30.0)
9 (52.9)
48 (33.3)
Being lower glucose level or hypoglycemic condition before the meal
38 (43.7)
3 (7.5)
2 (11.8)
43 (29.9)
Feel less worry of hypoglycemia if I inject after meal
6 (6.9)
6 (15.0)
3 (17.6)
15 (10.4)
Other
5 (5.7)
3 (7.5)
1 (5.9)
9 (6.3)
Ever not eaten meal soon after MTI injection, n (%)
Only patients who reported ever having not eaten meal soon after MTI injection were asked this question. Also, the percentage is reported for only the patients who responded in ‘Yes’, i.e., out of 41 (24.6%) patients in this case.
, what were the reasons for not eating meal soon after MTI injection, n (%)
Percentages do not add to 100 as patients may report more than one outcome. The denominators of the proportions are the number of patients with experience of dosing after a meal.
Due to physical condition
12 (29.3)
7 (24.1)
1 (10.0)
20 (25.0)
Due to change in appetite
9 (22.0)
13 (44.8)
3 (30.0)
25 (31.3)
Due to change of my schedule
14 (34.1)
9 (31.0)
3 (30.0)
26 (32.5)
Staying outside / working outside
8 (19.5)
10 (34.5)
3 (30.0)
21 (26.3)
Due to eating out
2 (4.9)
6 (20.7)
1 (10.0)
9 (11.3)
Due to sleeping
5 (12.2)
3 (10.3)
2 (20.0)
10 (12.5)
Other
7 (17.1)
2 (6.9)
3 (30.0)
12 (15.0)
Forgetting to inject MTI before start of meal in past week, n (%)
54 (32.3)
53 (37.1)
39 (54.2)
146 (38.2)
If Yes, number of times injected during/after meal in past week, Mean (SD)
Only patients who reported to have forgotten to inject MTI before start of meal in past week were asked this question. Also, the mean is reported for only the patients who responded in ‘Yes’, i.e., out of 54 (32.3%) patients in this case.
2.8 (10.8)
1.8 (3.5)
2.5 (3.8)
2.4 (7.1)
If Yes, number of times not injecting until next meal in past week, Mean (SD)
Only patients who reported to have forgotten to inject MTI before start of meal in past week were asked this question. Also, the mean is reported for only the patients who responded in ‘Yes’, i.e., out of 54 (32.3%) patients in this case.
1.0 (2.8)
1.7 (1.9)
1.1 (3.2)
1.3 (2.6)
Key: MTI: Mealtime Insulin; SD: Standard Deviation; T1D: Type 1 diabetes; T2D: Type 2 diabetes.
*Only patients who reported dosing during or after meal in the past month were asked this question. Also, the percentage is reported for only the patients who responded in ‘Yes’, i.e., out of 87 (57.1%) patients in this case
** Percentages do not add to 100 as patients may report more than one outcome. The denominators of the proportions are the number of patients with experience of dosing after a meal.
*** Only patients who reported ever having not eaten meal soon after MTI injection were asked this question. Also, the percentage is reported for only the patients who responded in ‘Yes’, i.e., out of 41 (24.6%) patients in this case.
**** Only patients who reported to have forgotten to inject MTI before start of meal in past week were asked this question. Also, the mean is reported for only the patients who responded in ‘Yes’, i.e., out of 54 (32.3%) patients in this case.
Concerning the percentage of the time patients dosed at different timings, dosing between 3 and 15 min and at the start of meal were the most frequently reported timings (36.7% and 36.4% of the time respectively) followed by dosing over 15 min before meal (15.9% of the time) and dosing after start of meal (11.0% of the time).
3.4 Healthcare provider instruction for MTI timings
Patients’ recall of healthcare provider (HCP) instruction for MTI injections is presented in Table 2. Patients were allowed to provide multiple responses to the questions related to the instructions provided by their healthcare provider regarding the MTI timings. In T1D and yT2D group, majority of patients (44.9% and 46.9% respectively) reported that their HCP instructed them to inject MTI at the start of meal, while in eT2D group, the most common HCP instruction was to inject between 3 and 15 min before the start of meal (reported by 47.2% patients). Fewer patients reported that their HCP instructed them to inject MTI over 15 min before the start of meal (14.4%, 11.2% and 16.7% patients in T1D, yT2D and eT2D group respectively), after the start of meal (4.2%, 4.2% and 2.8% patients in T1D, yT2D and eT2D group respectively) or gave no instruction at all (7.8%, 9.8% and 12.5% patients in T1D, yT2D and eT2D group respectively). Patterns of the combinations of responses to HCP instructed timings are provided in Supplemental Table S 2.
3.5 Reasons for having taken MTI injection during or after meal
The reasons reported by patients for injecting MTI during or after meal in past month are presented in Table 3. Amongst T1D patients who had dosed during or after meal in the past month, the most common reason was having lower glucose level or feeling hypoglycemic before the meal (43.7%), whereas amongst yT2D and eT2D patients who had dosed during or after meal in the past month, the most common reason was forgetting (42.5% and 64.7% respectively).
Lifestyle-related reasons such as eating outside of home (31.0%, 30.0% and 52.9% in T1D, yT2D and eT2D groups respectively), staying or working outside (21.8%, 15% and 29.4%), not able to find a place to inject (17.2%, 17.5% and 23.5%) and too busy to prepare MTI before meal (16.1%, 15% and 23.5%) were other frequently reported reasons for having dosed during or after meal in the last month. None of the elderly T2D patients reported reasons of eating more than planned or appetite fluctuations/uncertainty of how much to eat before a meal. Also, it was seen that more T2D patients (15.0% and 17.6% in yT2D and eT2D respectively) felt less worried of being hypoglycemic if they injected insulin after the meal, as compared to T1D patients (6.9%) (Table 3).
3.6 Missed dose of MTI injection
About a third of T1D patients and yT2D patients, and over half of eT2D patients reported forgetting to inject before the start of meal in the past week. Among the patients who missed an MTI injection before start of meal in the past week, the average number of times patients injected during/after meal was approximately 2.4 times, whereas the average number of times/week patients who reported not injecting until the next meal was approximately 1.3 times.
Regarding the dose adjustment by patients themselves, a total of 26.9%, 69.2% and 66.7% patients in T1D, yT2D and eT2D group respectively, reported that they took fixed doses of MTI as instructed by their HCP provider, while 73.1%, 30.8% and 33.3% patients respectively, reported adjusting their MTI dose according to their schedule, physical condition or basal glucose levels.
3.7 Not eating a meal soon after the injection
A total of 24.6%, 20.3% and 13.9% of T1D, yT2D and eT2D patients, respectively, reported not having eaten a meal at least once, soon after their MTI injection. Among those who reported not having eaten a meal soon after their MTI injection, the main reasons were change of schedule (34.1%, 31.0% and 30.0% in T1D, yT2D and eT2D groups, respectively), change of appetite (22.0%, 44.8% and 30.0%), staying or working outside (19.5%, 34.5% and 30.0%), and change in physical condition (29.3%, 24.1% and 10.0%). Also, 6 T1D patients reported in the free-text field that they skipped a meal because they had high blood glucose level before the start of meal.
3.8 Burden of current MTI injection timing
Burden related to the current MTI injection timings was assessed by the proportion of patients reporting moderate to severe burden for 12 different aspects (related to MTI intake), which they may perceive burdensome due to MTI timing (see Fig. 1 for details). Patients reported moderate or severe burden for a mean of 5.4 aspects out of the 12 (T1D: 6.1, yT2D: 5.1, eT2D: 4.5). In general, it was observed that more T1D patients reported moderate/severe burden than yT2D or eT2D patients.
Fig. 1Bar chart of reported burden of current MTI timing (including all burden items [moderate or severe response percentages]), by group (T1D, young T2D, elderly T2D),The order of questions is from highest to lowest percentage report of moderate or severe burden in the overall sample. The question numbers (Q1-Q11) reflect the order in which they were asked to the patients.
The aspect associated with the highest prevalence of moderate or severe burden of MTI timings was ‘Q6. Taking my insulin before the meal when I am out for social or leisure activities’, as reported by 75%, 62% and 67% of T1D, yT2D and eT2D patients respectively. Other aspects reported in decreasing order of their burden were ‘Q9. In general, securing a private place to inject before the meal when I’m out’, ‘Q7. Taking my insulin before the meal when my plan changes unexpectedly (e.g. invitation from friends or work colleagues, an urgent matter to attend to)’, ‘Q3. Because I don’t know the timing of food served, taking my insulin before the meal when eating out’, ‘Q12. I feel that injecting insulin before the meal restricts the scope of my activities’,‘Q10. I feel fear of experiencing hypoglycemia due to injecting insulin before the meal’, and ‘Q2. I don’t want to decide the amount to eat before eating, taking my insulin before the meal’.
3.9 Report of perceived change in amount of burden
The positive and negative perceived change in burden were assessed by asking patients ‘Imagine if your MTI worked faster and could be taken 2 min before or within 20 min after a start of meal. How do you think your opinion might change?’ for each statement. The responses of the patients with respect to perceived change in burden with a faster acting MTI is presented in Supplementary Figure S 2. In general, the majority of patients reported perceiving the burden to remain the same even if the current MTI was replaced with a faster acting MTI.
3.10 Factors associated with injection during or after meal
The factors associated with injection during or after meals was explored using a multivariable logistic regression analysis; the following variables were included: type of diabetes, HbA1c level, experience of severe hyperglycemia ever, experience of non-severe hyperglycemia in past 3 months, any symptom soon after meal in the past week, HCP instruction for injecting MTI after start of the meal, gender, age, educational status and full-time employment. Type of diabetes (T1D vs T2D odds ratio [OR]: 2.11, p = 0.003), experience of severe hypoglycemia (at least once in the last year vs never OR: 2.21, p = 0.013) and healthcare provider instruction for MTI timing (after start of meal vs not after start of meal OR: 4.93, p = 0.016) were found to have a significantly higher likelihood of MTI injection during or after a meal in the past month (Supplemental Table S 3).
4. Discussion
This study describes the real-world current MTI injection practice and highlights the challenges and unmet needs regarding MTI therapy in Japanese diabetic patients. Although, earlier studies have reported improper injection timing in diabetic patients [
], there is limited literature regarding this in the Japanese population. This study bridges this literature gap and additionally brings further insights on the factors associated with patients injecting MTI during or after meal and the action taken by them when they forget to inject their MTI dose.
In this study, it was observed that 43.1% of T1D, 38.5% of young T2D and 27.8% of elderly T2D patients reported normally injecting MTI at the start of a meal or ≤2 min before the start of a meal, which is not considered an ideal timing for post-prandial glucose control given MTI products currently available [
]. Moreover, although a small proportion of patients reported normally injecting MTI after the start of a meal (13.2%, 6.3% and 4.2% in T1D, yT2D and eT2D respectively), over half of T1D patients (52.1%) and about a quarter of yT2D and eT2D patients (28.0% and 23.6% respectively) reported having dosed during or after meal in the past month. Aside from normal timing and any report of dosing after meal, patients reported variability in the percentage of time they injected at certain times in relation to the meal. Specifically, dosing after start of meal was reported 11.0% of the time overall: 23.6% of the time for T1D patients, 29.4% of the time for yT2D patients and 15.6% of the time for eT2D patients. According to the prescribing information of the currently available MTI products in Japan, most products should be administered at least 15 min before meal for optimal post-prandial glucose control [
]. These findings underscore an unmet need related to the inflexible timing of the current MTI products available in Japan. More recently, a faster acting MTI has now been approved in Japan which can be administered at the start of the meal to within 20 min after starting the meal [
A small percentage of patients reported that HCPs instructed them to take MTI injections sufficiently before the start of a meal. Even though this data may be subject to recall bias, it is possible that this is reflective of the real-world prescribing practices in Japan where, in certain instances, HCPs may instruct patients to use currently available MTIs at timings other than the ideal prescription timing, considering their lifestyle and other conditions. The top five reasons reported by the patients (T1D and T2D) for not injecting MTI before meal were ‘Ate outside of home’, ‘Forgot to administer before meal’, ‘Being lower glucose level or hypoglycemic condition before the meal’, ‘Stayed outside/worked outside’ and ‘Couldn’t find a place to inject’. Similarly, moderate to strong burden of current MTI injection timing was consistently experienced by the diabetic patients as reported in the main aspects that were questioned viz., the burden felt while ‘taking insulin before meal when patient is out for social/leisure activities’, ‘securing a private place to inject’, ‘unexpected change in plans’ and ‘not knowing timing of food serving’. Taken together, these findings characterize the current burden for diabetic patients requiring MTI products, and therefore the need for other MTI alternatives to be made available to diabetic patients in Japan. For example, a formulation which achieves optimal post-prandial glucose control faster may help achieve better overall and post-prandial glucose control in Japanese patients who prefer to inject MTI at the start of a meal or after starting a meal.
When patients were asked to consider how their perceived burden may change if a fast-acting MTI was made available, a slightly greater percentage of patients reported ‘likely improve’ as compared to ‘likely worse’ for each statement. This is possibly because of default bias, which is due to the mindset of continuing with the current practice unless there is a critical issue; although, it also suggests that at least some patients wish for a fast-acting formulation that may enable them to enjoy a more active lifestyle. This, in turn, may help address some of the prevalent reasons for administrating MTI during/after the meal, such as being/eating outside or forgetting to inject insulin before a meal.
Forgetting to take the injection was the most common reason for dosing during or after meal in T2D patients (both young and elder), which is in line with previous studies [
]. T1D patients were more concerned about the dose titration issue, constantly accounting for their fear of experiencing hypoglycemia which, too, is in line with what has been reported in the literature [
]. In fact, feeling hypoglycemic before start of the meal was reported as a reason to avoid taking MTI injection by the T1D patients. Additionally, the T1D patients generally reported higher burden as compared to the T2D patients. This may be related to the greater number of hypoglycemic events experienced by T1D patients (as also reported in this study; Supplemental Table S 1). Previous studies have reported that T1D patients tend to have a lower quality of life than T2D patients [
Guided self-determination-young versus standard care in the treatment of young females with type 1 diabetes: study protocol for a multicentre randomized controlled trial.
]. However, other factors due to which T1D patients experience a higher burden than T2D patients were not assessed in this study and would require further research.
There were a few limitations associated with the present study. One of them was the recruitment of higher proportion of males than females, which may potentially lead to under-representation of females in the study; however, this higher proportion of males is consistent with the higher prevalence of diabetes in males in the Japanese population [
]. Another limitation was that many enrolled patients were already injecting MTI during or after a meal, and therefore may not have considered the risks of injecting at non-recommended timings in their perceptions of burden. Also, since there is no new fast-acting MTI formulation available in Japan, patients could only imagine a hypothetical situation while answering the survey and could only speculate the implications based on a short description and, therefore, might be subject to default bias. This may be because a patient may misjudge the burden if a question is asked to them long after he/she actually experience the event, leading to recall bias. This might also partially explain why the majority of patients responded that the burden would probably remain the same while comparing the perceived change in burden with the introduction of a new fast-acting MTI. A future study, which includes patients already experienced in using a fast-acting insulin product, is needed to better compare and evaluate the differences between injecting MTI before and after a meal.
5. Conclusion
In this study, overall 38.5% of patients reported that their normal MTI timing was at start of a meal or ≤2 min before start of a meal. Additionally, 8.9% of patients reported that their normal MTI timing was after the start of a meal, and 24.9% of patients reported injecting after start of meal in the past month. The MTI timings depended on the patient’s schedule, physical condition and blood glucose level. This timing of MTI dosing is not optimal to achieve adequate post-prandial glucose control by currently available therapies in Japan but maybe optimal for a faster acting therapy.
The main reason for dosing during or after meal in the past month among T1D patients was having lower glucose level or feeling hypoglycemic before the meal while among T2D patients was forgetting to inject before meal. Other frequent reasons reported by both T1D and T2D patients were lifestyle-related issues such as eating/working outside home or not being able to find a place to inject. Also, a notably high moderate to strong burden was reported by patients consistently in all key aspects that were questioned, most prominently when ‘out for social or leisure activities’, ‘securing a private place’, ‘my plan changes unexpectedly’ and ‘not knowing the timing of food served’. Physicians treating T1D and T2D patients in Japan should be aware of the extent of MTI usage at non-recommended timings, and the main reasons and burdens reported by patients of currently available MTI products.
Acknowledgement
The present study was sponsored by Eli Lilly and Company and it is in process of developing a faster acting mealtime insulin. The authors would like to acknowledge Sato, Takanori and Matsuda, Hiroyuki providing general administrative support for this study. The authors would also like to acknowledge Ishneet Kaur and Rajiv Sarkar (IQVIA, India) for their writing support, which was commissioned by Eli Lilly and Company.
Funding
This study was sponsored by Eli Lilly and Company.
Declaration of Competing Interest
TT, TA and RP are employees and a minor shareholder of Eli Lilly and Company. SS is a former employee of Eli Lilly Japan and a minor stakeholder of Eli Lilly and Company. HI has received research grants from Eli Lilly Japan and honoraria from Ono Pharmaceutical, Shionogi, Mitsubishi Tanabe Pharma, Arkray, MSD, Taisho Toyama Pharmaceutical, Becton–Dickinson Japan, Novo Nordisk Pharma, Daiichi Sankyo, Takeda Pharmaceutical, Astellas Pharma, and Boehringer Ingelheim Japan, and lecture and/or consultant fees from Takeda Pharmaceutical, Eli Lilly Japan, Sanofi, MSD, Astellas Pharma, Novartis Pharma, Mitsubishi Tanabe Pharma, Daiichi Sankyo, Ono Pharmaceutical, AstraZeneca, Taisho Toyama Pharmaceutical, Shionogi, Kowa Pharmaceutical, Boehringer Ingelheim Japan, Novo Nordisk Pharma, Sumitomo Dainippon Pharma, Kyowa Hakko Kirin, Terumo Co., Bristol-Myers Squibb Co., Sanwa Kagaku Kenkyusho Co. Ltd., Fujirebio Inc., Kowa Pharmaceutical, Pfizer, and Teijin Pharma. PW and SD are employees of IQVIA, the company commissioned by Eli Lilly and Company to conduct the project.
Appendix A. Supplementary data
The following are the Supplementary data to this article:
Duration of diabetes and types of diabetes therapy in Japanese patients with type 2 diabetes: The Japan diabetes complication and its prevention prospective study 3 (JDCP study 3).
The value of fast-acting insulin aspart compared with insulin aspart for patients with diabetes mellitus treated with bolus insulin from a UK health care system perspective.
Attitudes of patients and physicians to insulin therapy in Japan: an analysis of the global attitude of patients and physicians in insulin therapy study.
The current status of treatment-related severe hypoglycemia in Japanese patients with diabetes mellitus: A report from the committee on a survey of severe hypoglycemia in the Japan diabetes society.
Guided self-determination-young versus standard care in the treatment of young females with type 1 diabetes: study protocol for a multicentre randomized controlled trial.
31449-4&id=gr1.jpg){kind=link}