Reproducibility and Validity of a Questionnaire Measuring Treatment Burden on Patients with Type 2 Diabetes: Diabetic Treatment Burden Questionnaire (DTBQ)

Reproducibility and Validity of a Questionnaire Measuring Treatment Burden on Patients with Type... Diabetes Ther (2018) 9:1001–1019 https://doi.org/10.1007/s13300-018-0414-4 ORIGINAL RESEARCH Reproducibility and Validity of a Questionnaire Measuring Treatment Burden on Patients with Type 2 Diabetes: Diabetic Treatment Burden Questionnaire (DTBQ) . . . . Hitoshi Ishii Hiroki Shin Takahiro Tosaki Tatsuya Haga . . . Yoshiki Nakajima Toshihiko Shiraiwa Nobuaki Watanabe . . . . Miyuki Koizumi Hiroki Nakajima Sadanori Okada Tsuyoshi Mashitani Takako Mohri Yasuhiro Akai Received: January 16, 2018 / Published online: March 29, 2018 The Author(s) 2018 Treatment Burden Questionnaire (DTBQ), a ABSTRACT patient-administered questionnaire composed of 18 questions, and evaluated its reproducibil- Introduction: To measure the burden of phar- ity and validity. macotherapy on patients with type 2 diabetes Methods: We enrolled 240 patients with T2DM mellitus (T2DM), we developed the Diabetes under pharmacotherapy over 20 years of age at seven institutes in Japan. Their physicians filled Enhanced content To view enhanced content for this out report forms on patient backgrounds, and article go to https://doi.org/10.6084/m9.figshare. the patients answered both the DTBQ and the Diabetes Treatment Satisfaction Questionnaire Electronic supplementary material The online (DTSQ). For evaluation of reproducibility, 48 of version of this article (https://doi.org/10.1007/s13300- 018-0414-4) contains supplementary material, which is the enrolled subjects completed a 2nd DTBQ at available to authorized users. home after leaving the medical institutes. Results: Statistical analyses were performed for H. Ishii (&)  M. Koizumi  H. Nakajima  S. Okada two sets of subjects, the validity analysis set T. Mashitani  T. Mohri  Y. Akai (N = 236) and the reproducibility analysis set Department of Diabetology, Nara Medical (N = 47). Factor analysis found a simple struc- University, Nara, Japan e-mail: hits1@naramed-u.ac.jp ture in the DTBQ item scores using a three-fac- tor model with varimax rotation; the three H. Shin subscales were designated as ‘‘implementation Shin Clinic, Tokyo, Japan burden’’, ‘‘flexibility burden’’, and ‘‘blood glu- T. Tosaki cose control burden’’. All intraclass correlation TDE Healthcare Corporation TOSAKI Clinic for coefficients for the subscale scores were 0.8 or Diabetes and Endocrinology, Aichi, Japan higher, indicating high reproducibility. Nega- T. Haga tive correlations were observed between the Haga Diabetes Clinic, Aichi, Japan DTSQ satisfaction score and the DTBQ subscale scores. Moreover, as the dosing frequency of Y. Nakajima Nakajima Diabetes Clinic, Osaka, Japan diabetic medicines increased, the DTBQ total score (total burden score) also became higher. T. Shiraiwa Likewise, expected associations were observed Shiraiwa Medical Clinic, Osaka, Japan between patient backgrounds and DTSQ scores. N. Watanabe Watanabe Medical Clinic, Hyogo, Japan 1002 Diabetes Ther (2018) 9:1001–1019 Conclusion: The DTBQ has adequate repro- therefore have the greater risk associated with ducibility and validity as a measurement scale polypharmacy [10]. Considering the factors that for treatment burden on T2DM patients. might be causes of non-adherence, a less com- Trial Registration: University Hospital Medical plex and more convenient pharmacotherapy Information Network (UMIN) 000026382. would be preferable, and this together with a Funding: Eli Lilly Japan. reduction of treatment burden is a key for improving treatment. One possible solution for reducing treatment Keywords: Pharmacotherapy; Questionnaire; burden is the use of single-tablet fixed-dose Reproducibility; Treatment burden; Type 2 combinations of multiple OHAs [11, 12]. Fixed- diabetes mellitus; Validity dose combination therapies require fewer tablets to be taken and lessen treatment com- INTRODUCTION plexity, and thereby reduce treatment burden by replacing multiple agents with one fixed- Pharmacotherapy for type 2 diabetes mellitus dose combination. Another solution is the use (T2DM) has made remarkable progress since the of a long-acting formulation with a hypo- glycemic effect sustained for 1 week [13]. Cur- introduction of insulin, the first injectable treatment for diabetes, in the 1920s, rently, there are two available once-weekly (QW) formulations: an OHA DPP-4 inhibitor and sulfonylureas, the first oral hypoglycemic agent (OHA), in the 1950s [1]. Though the pro- and an injectable GLP-1 receptor agonist gress of pharmacotherapy has contributed to [14–17]. Some observational studies reported improving blood glucose (BG) control in T2DM that T2DM patients treated with QW GLP-1 patients, a substantial number of patients are receptor agonist exhibited better adherence still poorly controlled. In a survey of T2DM than patients treated with a once-daily (QD) adults in the USA, only 52.2% of the subjects agent [18, 19]. QW agents are largely equivalent achieved a glycemic target of HbA1c \ 7.0% in to the corresponding QD agents in terms of 2007–2010 [2]. Increasing the number of T2DM efficacy (i.e., hypoglycemic effect), safety (i.e., adverse events including hypoglycemia), and patients with a sufficient level of BG control is an important medical objective, and one way of cost (i.e., drug price). Therefore, if switching from QD to QW agents improved adherence, achieving this is to improve patient adherence to treatment. This is supported by a report which the reason would be that the treatment burden showed that a considerable portion of patients was reduced by mitigating the complexity of exhibited poor adherence [3] and some obser- QD treatment. However, there is currently no vational studies that also showed a positive questionnaire available for measuring quanti- correlation between adherence to medication tatively the difference of treatment burden and BG control level in the real world [4, 5]. between QW and QD agents. To measure burden on patients with dia- Multiple factors are considered to affect non- adherence to diabetic treatment [6–9]. Patient betes, two scales, PAID (Problem Areas in Dia- betes Survey) and DDS (Diabetes Distress Scale) factors include demographics (e.g., age, sex), psychological factors (e.g., health belief), phys- were developed so far [20–22]. However, these scales focus on the generic burden ical factors (e.g., concomitant disease), and social and economic status (e.g., education, attributable to diabetes, not on the treatment income). On the other hand, there are specific burden caused by pharmacotherapy. Therefore, pharmacotherapy factors: efficacy, safety, cost, they are not suitable for differentiating medici- complexity (e.g., frequency and interval of nes according to their characteristics, such as dose), and convenience (e.g., ease of swallowing dosing schedule which dominantly determines formulation, ease of handling injection device). the complexity and convenience of implemen- tation of pharmacotherapy. Thus, to measure In addition, T2DM patients, especially older adults, commonly have concomitant diseases, the treatment burden caused by pharma- cotherapy for T2DM, we developed DTBQ such as hypertension and dyslipidemia, and Diabetes Ther (2018) 9:1001–1019 1003 (Diabetes Treatment Burden Questionnaire) • Injection of GLP-1 receptor agonist, once weekly ± OHA (Injection/QW) focusing on efficacy (i.e., BG control), safety (i.e., hypoglycemia), complexity (i.e., amount, • Injection of insulin or GLP-1 receptor ago- nist, once daily ± OHA (Injection/QD) frequency, timing, and interval of dose), and convenience (i.e., ease of taking or injecting • Injection of insulin or GLP-1 receptor ago- medicine), and evaluated its reproducibility and nist, twice a day or more ± OHA (Injection/ validity. This questionnaire would make it pos- BID?) sible to quantify the treatment burden on • Only OHA, once weekly (OHA/QW) T2DM patients depending on the characteristics • Only OHA, once daily (OHA/QD) of medicines. • Only OHA, twice a day or more (OHA/BID?) In addition, subjects had to be at least 20 years old and provide written informed METHODS consent. Moreover, the following patients were excluded: patients who were diagnosed with or Development of DTBQ suspected to have dementia, patients with psy- chiatric disorders, patients without sufficient We interviewed 16 patients with T2DM treated judgment ability, patients who needed consent with pharmacotherapy about treatment burden. from their proxies, and other patients judged to Based on the interview, pre-existing question- be inappropriate by the attending physicians. naires, and opinions from diabetologists, the first The target numbers of enrolled patients were version of questionnaire with 26 questions was the following: 210 subjects for validity evalua- drafted. This first version was completed by 20 tion who answered DTBQ at least once, and 70 patients for pilot testing. Based on these results, subjects for reproducibility evaluation who 18 questions were selected for the final version. answered DTBQ twice. The enrollment was Each question is answered by choosing one of adjusted to balance the number of patients seven ordered categories: (1) strongly disagree, among the different types of diabetic treatment. (2) mostly disagree, (3) slightly disagree, (4) Patients enrolled as diabetic treatment includ- neither agree nor disagree, (5) slightly agree, (6) ing injectable medicines had the option to use mostly agree, (7) strongly agree. It should be OHAs together, so they were asked to answer noted that the relations between the category the questions in DTBQ on the number and level of treatment burden are injectable medicines only. opposite for questions 1–10 and questions The attending physician gave the ‘‘Report 11–18. In other words, a larger category number form for all patients’’ along with the 1st DTBQ means heavier treatment burden in questions and DTSQ to be filled out to the patient after 1–10, but a smaller burden in questions 11–18. confirming his/her eligibility; the ‘‘Report form Therefore, item scores are defined by the cate- for reproducibility evaluation’’, which included gory number converted so that 0 means ‘‘mini- the 2nd DTBQ, was then delivered only if the mum treatment burden’’ and 6 means patient was a subject for reproducibility evalu- ‘‘maximum treatment burden’’. Correspondence ation. The patient filled out the ‘‘Report form between the category number and the item score for all patients’’ without the presence of the for each question of DTBQ is shown in Table S1 attending physician in a room at the medical in the electronic supplementary material. institute. Then the patient put the form in an envelope and posted it to the data center or gave it to the attending physician. The attend- Evaluation of Reproducibility and Validity ing physician filled out the ‘‘Report form for physicians’’ and posted it to the data center All subjects were outpatients with T2DM treated with the ‘‘Report form for all patients’’ if he had for at least 12 weeks with pharmacotherapy received it from the patient. Patients who were corresponding to one of the following six types subjects for reproducibility evaluation filled out of diabetic treatment: the ‘‘Report form for reproducibility evaluation’’ 1004 Diabetes Ther (2018) 9:1001–1019 and posted it to the data center after leaving the for the validity analysis set. Based on the results, medical institute. subscales were determined and named. Intra- DTSQ is the most popular patient-reported class correlation coefficients (ICC) were calcu- questionnaire to evaluate patient satisfaction in lated for the DTBQ subscale and total scores for diabetic treatment [23]. License and copyright the reproducibility analysis set. Correlation agreement for using DTSQ questionnaire in this coefficients of the DTBQ subscale and total study was obtained from Health Psychology scores and the DTSQ scores were calculated for Research Limited (University of London, UK). the validity analysis set. In addition, analysis of The study was conducted from November 2016 variance (ANOVA) or the t test was used to to February 2017 at seven sites in Japan and was examine associations between DTBQ subscale registered at University Hospital Medical Infor- and total scores and patient background factors. mation Network (UMIN000026382). A common protocol was approved at insti- RESULTS tutional review boards in Nara Medical Univer- sity and TDE Healthcare Corporation TOSAKI Patients Clinic for Diabetes and Endocrinology. All pro- cedures performed in studies involving human Patient disposition is shown in Fig. 1. Out of participants were in accordance with the ethical standards of the institutional and/or national 240 enrolled patients, 236 were included in the validity analysis set; four were excluded because research committee and with the 1964 Decla- ration of Helsinki and its later amendments or of unanswered questions in the 1st DTBQ. The numbers of patients by type of diabetic treat- comparable ethical standards. Informed con- sent was obtained from all individual partici- ment were as follows: (1) Injection/QW, 40; (2) Injection/QD, 41; (3) Injection/BID?, 37; (4) pants included in the study. OHA/QW, 38; (5) OHA/QD, 35; and (6) OHA/ BID?, 45. Out of 236 patients in the validity Statistical Analyses analysis set, 48 patients filled out the 2nd DTBQ. One patient was excluded because of Enrolled patients who left any unanswered unanswered questions, and 47 were included questions in the 1st DTBQ were excluded from the reproducibility analysis set. the validity analysis set. Of the validity analysis Patient backgrounds reported by physicians set, patients who completed the 2nd DTBQ and patients are summarized in Table 1 and collected and without any unanswered ques- Table S2 (in the electronic supplementary tions were included in the reproducibility material), respectively. All background factors analysis set. Appropriate descriptive statistics were similarly distributed in the validity and were calculated for background factors, such as reproducibility analysis sets. type of diabetic treatment, HbA1c, age, and sex, for the validity and reproducibility analysis sets. Summary and Reproducibility of DTBQ Mean and standard deviation were calculated Item Scores for DTBQ item scores, and for the validity and reproducibility analysis sets. In addition, weighted kappa coefficients were calculated for Descriptive statistics of the DTBQ item scores for the validity and reproducibility analysis sets, the reproducibility analysis set where the weights (w) were assigned as follows: exact and weighted kappa coefficients for the repro- agreement, w = 1; one-level disagreement, ducibility analysis set are shown in Table 2. For all questions, item scores of 0 to 2, which indi- w = 5/6; two-level disagreement, w = 4/6; three- level disagreement, w = 3/6; four-level dis- cate a relatively small burden, were reported from more than half of the patients in the agreement, w = 2/6; five-level disagreement, w = 1/6; six-level disagreement, w = 0/6. Factor validity analysis set. The averages of 14 item scores out of the 18 questions were less than 2.0. analysis of the DTBQ item scores was conducted Diabetes Ther (2018) 9:1001–1019 1005 Fig. 1 Patient disposition. Full description of type of agent, once a week; 5. Oral hypoglycemic agent, once a day; diabetic treatment: 1. Injection of GLP-1 agonist, once a 6. Oral hypoglycemic agent, twice a day or more. BID? week ± oral hypoglycemic agent; 2. Injection of insulin or twice a day or more, DTBQ Diabetic Treatment Burden GLP-1 agonist, once a day ± oral hypoglycemic agent; 3. Questionnaire, OHA oral hypoglycemic agent, QD once Injection of insulin or GLP-1 agonist, twice a day or daily, QW once weekly more ± oral hypoglycemic agent; 4. Oral hypoglycemic Four questions whose average scores were medication (j = 0.550)’’, ‘‘8. Concern about greater than 2.0 were as follows: ‘‘7. Medication hypoglycemia (j = 0.532)’’, ‘‘17. Feeling less away from home’’, ‘‘13. Allowing me to take a burden to continue diabetes treatment missed dose’’, ‘‘16. Feeling less burden to follow (j = 0.560)’’, and ‘‘18. Satisfaction with my diet therapy’’, and ‘‘18. Satisfaction with my current BG control (j = 0.579)’’, which indi- current BG control’’. cated reproducibility good enough to use in Means of the 1st and 2nd item scores in the practice. reproducibility analysis set were close to those in the validity analysis set. Weighted kappa Factor Analysis of DTBQ Item Scores coefficients for the reproducibility analysis set were estimated at 0.6 or higher for 14 of the 18 Table 3 shows the results of factor analysis of questions. Four questions with point estimates the DTBQ item scores for the validity analysis less than 0.6 were ‘‘3. Securing time for 1006 Diabetes Ther (2018) 9:1001–1019 Table 1 Patient backgrounds reported by physicians Background factors Validity analysis set Reproducibility analysis set Number of patients 236 (100.0%) 47 (100.0%) Type of diabetic treatment Injection/QW 40 (16.9%) 10 (21.3%) Injection/QD 41 (17.4%) 8 (17.0%) Injection/BID+ 37 (15.7%) 5 (10.6%) OHA/QW 38 (16.1%) 7 (14.9%) OHA/QD 35 (14.8%) 5 (10.6%) OHA/BID+ 45 (19.1%) 12 (25.5%) HbA1c (%), mean (SD) 6.98 (0.92) 7.28 (1.14) Age (years), mean (SD) 63.4 (11.9) 62.8 (12.0) Sex Male 142 (60.2%) 27 (57.4%) Female 94 (39.8%) 20 (42.6%) Duration of diabetes (years), mean (SD) 13.1 (9.6) 13.5 (12.4) Frequency of dose for injection None 118 (50.0%) 24 (51.1%) Once a week 40 (16.9%) 10 (21.3%) Once a day 41 (17.4%) 8 (17.0%) Twice a day or more 37 (15.7%) 5 (10.6%) Type of injection Insulin 68 (28.8%) 13 (27.7%) GLP-1 agonist (QD, BID) 12 (5.1%) 0 (0.0%) GLP-1 agonist (QW) 40 (16.9%) 10 (21.3%) Frequency of dose for OHA None 21 (8.9%) 3 (6.4%) Once a week 40 (16.9%) 7 (14.9%) Once a day 62 (26.3%) 11 (23.4%) Twice a day 48 (20.3%) 9 (19.1%) Three times a day or more 65 (27.5%) 17 (36.2%) Type of OHA Sulfonylureas 61 (25.8%) 15 (31.9%) Glinide 39 (16.5%) 6 (12.8%) DPP-4 inhibitor 128 (54.2%) 25 (53.2%) Diabetes Ther (2018) 9:1001–1019 1007 Table 1 continued Background factors Validity analysis set Reproducibility analysis set Thiazolidine derivative 24 (10.2%) 4 (8.5%) Biguanide 81 (34.3%) 18 (38.3%) a-Glucosidase inhibitor 45 (19.1%) 14 (29.8%) SGLT2 inhibitor 30 (12.7%) 8 (17.0%) Diabetic neuropathy No 147 (62.3%) 27 (57.4%) Yes 52 (22.0%) 15 (31.9%) Unknown 37 (15.7%) 5 (10.6%) Diabetic nephropathy No 174 (73.7%) 30 (63.8%) Yes 62 (26.3%) 17 (36.2%) Diabetic retinopathy No 150 (63.6%) 30 (63.8%) Yes 61 (25.8%) 16 (34.0%) Unknown 25 (10.6%) 1 (2.1%) BID+ twice a day or more, OHA oral hypoglycemic agent, QD once daily, QW once weekly set. In the factor pattern for a four-factor model Reproducibility of DTBQ Subscale Scores without rotation, the first factor loadings were and Total Burden Score more than 0.4 for all questions, which indicated a unidimensional feature of DTBQ. Therefore, Scatter plots of the DTBQ subscale scores and the total score of all 18 questions was desig- total burden score (1st vs. 2nd) and the esti- nated as the ‘‘total burden score’’. On the other mates of ICC for the reproducibility analysis set hand, the variance of three factors was greater are shown in Fig. 2. The ICC estimates were 0.8 than 1.0 in the four-factor model. or higher, indicating high reproducibility of all In the factor pattern for three-factor model scores. with varimax rotation, focusing on factor load- ing more than 0.4, a simple structure was Correlation Between DTBQ and DTSQ obtained, where all questions belonged to one Scores of the three factors. Considering the factor pattern and contents of the questions, three Table 4 shows the univariate descriptive statis- subscales were determined and designated: tics and correlations of the DTBQ and DTSQ • Implementation burden score: sum of item scores for the validity analysis set. Correlations scores 1–10 between the DTBQ subscale scores were 0.4 or • Flexibility burden score: sum of item scores higher for Spearman’s correlation coefficient 11–13 (SCC) for every combination. The DTBQ total • Blood control (BG) control burden score: burden scores correlated positively with all the sum of item scores 14–18 1008 Diabetes Ther (2018) 9:1001–1019 Table 2 Reproducibility of DTBQ item scores (reproducibility analysis set) Questions abbreviated Reproducibility analysis set (N = 47) 1st mean 2nd mean Weighted j coefficient (SD) (SD) Point 95% confidence estimate interval 1. Medication on time 1.7 (1.6) 1.5 (1.6) 0.785 (0.665, 0.906) 2. Medication during busy hours 1.7 (1.7) 1.6 (1.6) 0.807 (0.710, 0.904) 3. Securing time for medication 0.9 (1.1) 1.1 (1.2) 0.550 (0.355, 0.744) 4. Pain associated with medication 1.3 (1.8) 1.3 (1.7) 0.690 (0.538, 0.841) 5. Feeling that I should not miss a dose 1.8 (1.7) 1.5 (1.5) 0.697 (0.549, 0.846) 6. Feeling guilty when I miss a dose 1.7 (1.9) 1.6 (1.8) 0.839 (0.752, 0.926) 7. Medication away from home 1.9 (1.8) 2.0 (1.9) 0.799 (0.705, 0.892) 8. Concern about hypoglycemia 0.9 (1.2) 1.1 (1.2) 0.532 (0.333, 0.731) 9. Inflexibility to adjust the time for medication 1.0 (1.3) 1.2 (1.3) 0.677 (0.505, 0.848) 10. Worrying about future 1.8 (1.7) 1.9 (1.7) 0.710 (0.579, 0.840) 11. Short time and small effort for medication 1.1 (1.5) 1.1 (1.2) 0.627 (0.434, 0.821) 12. Medication without time pressure 2.0 (1.8) 1.9 (1.9) 0.811 (0.702, 0.920) 13. Allowing me to take a missed dose 1.9 (2.1) 1.9 (1.9) 0.791 (0.693, 0.889) 14. Feeling that my diabetes is getting better 1.9 (1.3) 2.0 (1.3) 0.713 (0.586, 0.839) 15. Allowing me to control BG with small effort 1.8 (1.3) 1.7 (1.3) 0.650 (0.484, 0.816) 16. Feeling less burden to follow diet therapy 2.0 (1.5) 2.1 (1.5) 0.643 (0.481, 0.804) 17. Feeling less burden to continue diabetes treatment 1.5 (1.4) 1.7 (1.3) 0.560 (0.392, 0.728) 18. Satisfaction with my current BG control 2.0 (1.5) 2.1 (1.7) 0.579 (0.414, 0.745) Range of item scores: 0 (minimum treatment burden)–6 (maximum treatment burden) DTBQ subscale scores with 0.7 or higher SCC Association Between DTBQ Scores estimates. and Patient Backgrounds The DTSQ hyperglycemic score correlated positively with the DTBQ scores: implementa- For the validity analysis set, analysis results of tion burden (SCC = 0.31), BG control burden the association between DTBQ scores and (0.36), and total burden (0.36). Weak correla- patient backgrounds reported by physicians and tions were observed between the DTSQ hypo- patients are shown in Table 5 and Table S3 (in glycemic score and the DTBQ scores within the the electronic supplementary material), respec- range of ± 0.2 SCC. The DTSQ satisfaction score tively. In addition, the results of pairwise com- correlated negatively with all the DTBQ scores parisons of the DTBQ/DTSQ scores between with - 0.3 SCC or less. In particular, the BG types of diabetic treatment are shown in control burden score (SCC = - 0.61) and total Table S4 in the electronic supplementary burden score (- 0.53) were correlated strongly. material. Diabetes Ther (2018) 9:1001–1019 1009 Table 3 Factor analysis of DTBQ item scores (validity analysis set) Questions abbreviated 4-factor model without rotation 3-factor model with varimax rotation Factor loading Factor loading Communality 12 3 4 12 3 1. Medication on time 0.743* - 0.321 - 0.004 - 0.124 0.755* 0.164 0.112 0.609 2. Medication during busy 0.734* - 0.393 0.003 0.006 0.817* 0.105 0.081 0.685 hours 3. Time ensure for medication 0.645* - 0.329 0.022 0.300 0.694* 0.095 0.097 0.500 4. Pain associated with 0.408* - 0.293 - 0.164 0.449* 0.479* 0.088 - 0.088 0.245 medication 5. To feel that I should not 0.705* - 0.384 - 0.036 - 0.137 0.803* 0.112 0.044 0.659 miss a dose 6. To feel guilty when I miss a 0.464* - 0.276 0.110 - 0.334 0.527* - 0.004 0.118 0.291 dose 7. Medication away from home 0.711* - 0.248 0.043 - 0.105 0.720* 0.162 0.169 0.573 8. Concern about 0.503* - 0.108 0.023 0.210 0.462* 0.163 0.144 0.261 hypoglycemia 9. Not to flexibly adjust the 0.664* - 0.242 0.086 - 0.092 0.675* 0.122 0.188 0.506 time for medication 10. To worry about future 0.604* - 0.190 - 0.093 - 0.048 0.591* 0.219 0.052 0.400 11. Short time and small effort 0.408* 0.331 0.326 0.260 0.124 0.245 0.530* 0.356 for medication 12. Medication without time 0.487* 0.426* 0.522* - 0.033 0.124 0.236 0.796* 0.704 pressure 13. To allow me to take a 0.427* 0.377 0.510* - 0.058 0.106 0.180 0.741* 0.592 missed dose 14. To feel that my diabetes is 0.422* 0.520* - 0.218 0.023 0.025 0.676* 0.189 0.493 getting better 15. To allow me to control BG 0.578* 0.528* - 0.243 - 0.043 0.144 0.780* 0.222 0.678 with small effort 16. To feel less burden to 0.616* 0.454* - 0.146 - 0.025 0.219 0.694* 0.284 0.611 follow diet therapy 17. To feel less burden to 0.611* 0.422* - 0.134 0.117 0.235 0.659* 0.280 0.568 continue diabetes treatment 18. Satisfied with my current 0.511* 0.391 - 0.423* - 0.165 0.180 0.730* 0.001 0.566 BG control Variance of factors 6.061 2.361 1.023 0.619 4.606 2.847 1.846 1010 Diabetes Ther (2018) 9:1001–1019 Table 3 continued Questions abbreviated 4-factor model without rotation 3-factor model with varimax rotation Factor loading Factor loading Communality 12 3 4 12 3 Cronbach’s a 0.885 0.887 0.862 0.775 N = 236. *Factor loading is more than 0.4 or less than - 0.4 Fig. 2 Reproducibility of DTBQ subscale scores and total 11–13), c BG control burden score (sum of scores for burden scores (reproducibility analysis set). a Implementa- items 14–18), d total burden score (sum of scores for items tion burden score (sum of scores for items 1–10), 1–18). CI confidence interval b flexibility burden score (sum of scores for items DTBQ implementation burden score had dose: injection and OHA, diabetic retinopathy) significant associations with three items repor- and three items reported by patients (i.e., ted by physicians (i.e., type of diabetic treat- hypoglycemic event, frequency of missed dose: ment, age, frequency of dose: OHA) and five injection or OHA, time to leave home on items reported by patients (i.e., BG control, weekdays). DTBQ flexibility burden score was hypoglycemic event, frequency of missed dose: high in the following patients: frequently dos- injection, OHA, injection or OHA). DTBQ ing with injection or OHA, higher HbA1c implementation burden scores were high in the (C 7.0%), older age (C 65 years), longer dura- following patients: treated with injection, fre- tion of diabetes (C 10 years), with diabetic quently dosing with injection or OHA, younger retinopathy, with hypoglycemic events, fre- age (\ 65 years), with BG controlled poorly, quently missing dose of injection or OHA, with hypoglycemic events, frequently missing leaving home on weekdays late (8:30 or later). dose of injection or OHA. DTBQ BG control burden score had signifi- DTBQ flexibility burden score had significant cant associations with four items reported by associations with seven items reported by physicians (i.e., type of diabetic treatment, physicians (i.e., type of diabetic treatment, HbA1c, diabetic nephropathy and retinopathy) HbA1c, age, duration of diabetes, frequency of and four items reported by patients (i.e., BG Diabetes Ther (2018) 9:1001–1019 1011 Table 4 Correlation between DTBQ scores and DTSQ scores (validity analysis set) Univariate statistics DTBQ DTSQ Implementation Flexibility BG control burden Total Hyperglycemia Hypoglycemia Satisfaction burden burden burden Mean 13.5 5.3 9.3 28.1 2.5 1.1 27.4 Standard deviation 11.6 4.9 6.0 17.5 1.8 1.5 6.1 N 236 236 236 236 236 236 236 Minimum 0 0 0 0 0 0 12 Median 10.0 5.0 9.0 27.0 2.5 0.0 28.5 Maximum 51 18 26 76 6 6 36 Correlation coefficient matrix: Spearman’s correlation coefficient DTBQ Implementation 1.00 – – – – – – burden Flexibility burden 0.43 1.00 – – – – – BG control burden 0.42 0.46 1.00 – – – – Total burden 0.86 0.70 0.73 1.00 – – – DTSQ Hyperglycemia 0.31 0.16 0.36 0.36 1.00 – – Hypoglycemia 0.14 0.10 0.03 0.12 0.25 1.00 – Satisfaction - 0.38 - 0.36 - 0.61 - 0.53 - 0.31 - 0.07 1.00 BG blood glucose, DTBQ Diabetic Treatment Burden Questionnaire, DTSQ Diabetic Treatment Satisfaction Questionnaire 1012 Diabetes Ther (2018) 9:1001–1019 Table 5 Associations between DTBQ scores and patient backgrounds reported by physicians (validity analysis set) Patients’ backgrounds N DTBQ scores Implementation burden Flexibility burden BG control burden Total burden Category Mean (SD) Mean (SD) Mean (SD) Mean (SD) Validity analysis set 236 13.5 (11.6) 5.3 (4.9) 9.3 (6.0) 28.1 (17.5) Type of diabetic treatment Injection/QW 40 12.3 (11.0) 2.7 (3.5) 9.4 (5.4) 24.3 (14.9) Injection/QD 41 18.3 (13.9) 6.6 (4.3) 9.3 (6.1) 34.2 (18.3) Injection/BID? 37 17.5 (13.2) 8.1 (5.4) 11.1 (6.7) 36.6 (20.4) OHA/QW 38 7.9 (8.3) 2.0 (2.7) 7.0 (4.3) 17.0 (12.0) OHA/QD 35 8.2 (7.0) 4.6 (4.5) 8.4 (6.8) 21.1 (12.9) OHA/BID? 45 15.8 (10.1) 7.6 (5.0) 10.6 (6.1) 33.9 (15.8) ANOVA P \ 0.001*** P \ 0.001*** P = 0.044* P \ 0.001*** HbA1c \ 7.0% 133 12.6 (11.6) 4.7 (4.5) 7.9 (5.3) 25.2 (16.6) 7.0% or higher 103 14.6 (11.5) 6.1 (5.3) 11.1 (6.5) 31.9 (17.9) t test P = 0.186 P = 0.022* P \ 0.001*** P = 0.004** Age \65 115 15.1 (11.5) 4.6 (4.1) 9.6 (5.9) 29.3 (17.1) 65 or older 121 12.0 (11.5) 6.0 (5.5) 9.0 (6.2) 27.0 (17.8) t test P = 0.038* P = 0.031* P = 0.486 P = 0.310 Sex Male 142 13.1 (11.3) 5.6 (5.1) 9.3 (6.0) 27.9 (17.0) Female 94 14.2 (12.0) 4.9 (4.6) 9.4 (6.2) 28.4 (18.2) t test P = 0.476 P = 0.255 P = 0.841 P = 0.824 Duration of diabetes \ 10 years 93 13.2 (11.9) 4.4 (4.4) 9.1 (5.5) 26.7 (17.7) 10 years or longer 142 13.8 (11.4) 5.8 (5.1) 9.4 (6.3) 29.0 (17.3) Unknown 1 7.0 (–) 15.0 (–) 25.0 (–) 47.0 (–) t test (unknown excluded) P = 0.698 P = 0.030* P = 0.732 P = 0.328 Frequency of dose: injection None 118 11.0 (9.4) 4.9 (4.8) 8.8 (6.0) 24.6 (15.6) 1/week 40 12.3 (11.0) 2.7 (3.5) 9.4 (5.4) 24.3 (14.9) 1/day 41 18.3 (13.9) 6.6 (4.3) 9.3 (6.1) 34.2 (18.3) 2/day 8 10.4 (9.4) 7.1 (5.0) 9.4 (5.7) 26.9 (13.9) Diabetes Ther (2018) 9:1001–1019 1013 Table 5 continued Patients’ backgrounds N DTBQ scores Implementation burden Flexibility burden BG control burden Total burden Category Mean (SD) Mean (SD) Mean (SD) Mean (SD) 3/day 8 19.5 (13.3) 8.4 (6.1) 14.1 (8.2) 42.0 (25.0) 4/day 21 19.4 (13.9) 8.4 (5.5) 10.5 (6.5) 38.3 (20.3) ANOVA (none excluded) P = 0.082 P \ 0.001*** P = 0.306 P = 0.011* Frequency of dose: OHA None 21 15.9 (13.4) 8.7 (5.2) 10.8 (5.9) 35.4 (19.3) 1/week 40 9.4 (10.2) 2.2 (2.8) 7.3 (4.4) 18.9 (14.4) 1/day 62 11.9 (11.3) 5.0 (4.8) 9.1 (6.8) 25.9 (16.8) 2/day 48 13.8 (11.0) 5.4 (4.6) 9.2 (6.2) 28.4 (17.7) 3/day or more 65 16.6 (11.6) 6.4 (5.1) 10.4 (5.9) 33.4 (16.5) ANOVA (none excluded) P = 0.009** P \ 0.001*** P = 0.093 P \ 0.001*** Diabetic neuropathy No 147 13.0 (11.1) 4.8 (4.7) 9.0 (5.8) 26.8 (16.4) Yes 52 14.6 (12.8) 5.9 (4.9) 9.4 (6.1) 30.0 (18.9) Unknown 37 14.0 (11.9) 6.4 (5.6) 10.4 (6.9) 30.7 (19.2) t test (unknown excluded) P = 0.379 P = 0.154 P = 0.681 P = 0.259 Diabetic nephropathy No 174 13.1 (11.4) 5.0 (4.9) 8.8 (6.1) 26.9 (17.2) Yes 62 14.7 (12.1) 6.2 (5.0) 10.7 (5.8) 31.6 (17.9) t test P = 0.355 P = 0.081 P = 0.039* P = 0.069 Diabetic retinopathy No 150 13.0 (11.9) 4.8 (4.7) 8.7 (5.8) 26.4 (17.3) Yes 61 14.5 (10.0) 6.7 (4.9) 10.7 (6.1) 31.9 (16.4) Unknown 25 14.2 (13.3) 5.2 (5.4) 9.8 (6.7) 29.2 (20.0) t test (unknown excluded) P = 0.378 P = 0.007** P = 0.031* P = 0.036* ANOVA analysis of variance, BG blood glucose, BID? twice a day or more, DTBQ Diabetic Treatment Burden Ques- tionnaire, OHA oral hypoglycemic agent, QD once daily, QW once weekly *P \ 0.05, **P \ 0.01, ***P \ 0.001 control, hypoglycemic event, frequency of injection or OHA, higher HbA1c (C 7.0%), with missed dose: injection or OHA, communication diabetic nephropathy or retinopathy, with BG with physician on diabetic treatment). DTBQ controlled poorly, with hypoglycemic events, BG control burden score was high in the fol- frequently missing dose of injection or OHA, lowing patients: frequently missing dose of less communication with their physicians. 1014 Diabetes Ther (2018) 9:1001–1019 treatment. In particular, the flexibility burden DISCUSSION score rose significantly as the frequency of dose increased. Implementation burden score also We developed DTBQ, a patient-administered increased with order of dosing frequency, and questionnaire with 18 questions, to measure the was higher for injection than OHA. treatment burden caused by pharmacotherapy HbA1c was significantly associated with total for T2DM. The reproducibility and validity of burden score. Total burden score was higher in DTBQ were evaluated through statistical analy- patients with higher HbA1c (C 7.0%). Of the sis of data collected using the questionnaire DTBQ subscales, BG control burden score and from patients with T2DM under pharma- flexibility burden score also showed significant cotherapy. This included 236 patients as the differences with HbA1c. BG control reported by validity analysis set and 47 patients as the patients was significantly associated with DTBQ reproducibility analysis set. scores as well. These results suggested that Weighted kappa coefficients were calculated patients with worse BG control felt a heavier as an index of reproducibility for the DTBQ item burden for treatment of diabetes. scores, with all 18 questions showing repro- Age was not significantly associated with ducibility high enough for practical use. ICCs total burden score. However, among the DTBQ were also calculated as an index of repro- subscales, implementation burden score was ducibility for the DTBQ subscale scores and significantly higher in younger patients total burden score. All coefficients were 0.8 or (\ 65 years). Younger patients might feel heav- higher, showing high reproducibility of three ier implementation burdens because of their subscales and total burden score. greater job demands. On the other hand, flexi- Factor analysis of the DTBQ item scores bility burden score was significantly higher in without rotation showed a unidimensional older patients (C 65 years). feature of DTBQ, which supported interpreting Hypoglycemic events were significantly the total score of all 18 questions as ‘‘total bur- associated with total burden score and DTBQ den’’. On the basis of the result of the three- subscale scores. All scores were significantly factor model with varimax rotation, a simple higher in patients with hypoglycemic events. structure was obtained, and three subscales were Out of the six types of diabetic treatment, the determined and designated as follows: ‘‘imple- percentage of hypoglycemic events and all mentation burden’’, ‘‘flexibility burden’’, and DTBQ subscale scores were highest for Injec- ‘‘blood control (BG) control burden’’. tion/BID?, which reflected the association There were significant associations between between hypoglycemia and DTBQ scores. This type of diabetic treatment and DTBQ total bur- suggests that patients with hypoglycemic events den score. In terms of administration route would probably feel a heavier treatment burden (injection vs. OHA), the score was higher for for controlling BG. injection than oral administration. For the fre- Daily frequency of dosing non-diabetic quency of dose (QW, QD, BID?), the score agents was not significantly associated with any increased with order of dosing frequency: QW, DTBQ score. This result suggested that DTBQ QD, and BID?. Comparing the means of total measured the treatment burden caused by burden score among the type of diabetic treat- pharmacotherapy for diabetes only, as ment (combination of administration route and intended. dose frequency), OHA/BID? (33.9) was almost Frequency of missed dose of diabetic medi- the same as Injection/QD (34.2), and Injection/ cation (injection or OHA) was significantly QW (24.3) was smaller than OHA/BID?. These associated with the total burden score and all results suggested that the treatment burden of DTBQ subscale scores. All scores were signifi- Injection/QW was lighter than that of OHA/ cantly higher in patients with a missed injec- BID?. tion or OHA dose. As the frequency of dosing The DTBQ subscale scores were also signifi- diabetic medication increased, the frequency of cantly associated with type of diabetic missed dose and all subscale scores also Diabetes Ther (2018) 9:1001–1019 1015 increased simultaneously, which reflected the clearly preferred to QD. The questions in DTBQ association between the frequency of missed directly address factors affecting patients’ pref- dose and the DTBQ scores. erence, so DTBQ can contribute to a better The treatment burden in patients with dia- understanding of patient preference. betic retinopathy was higher than that in This study has some limitations. In the patients without diabetic retinopathy. In the validity analysis set of this study, mean HbA1c patients with diabetic retinopathy, 41 out of 61 was 6.98, which indicated that the study sub- subjects had injection therapy compared with jects controlled BG better than the general 63 out of 150 subjects in the patients without T2DM population. Another study should be diabetic retinopathy. Duration of illness of conducted to evaluate the validity of DTBQ for patients with diabetic retinopathy is relatively poorly controlled T2DM patients. Because the longer and insulin therapy is necessary for type of diabetic treatment was not randomized, many patients. Since injection therapy has a results based on comparisons of DTBQ scores higher burden overall, there was a statistically with treatment type should not be interpreted significant difference between the groups. as interventional effects. The treatment burden DTSQ satisfaction score was negatively cor- should be measured from various points of related with all DTBQ scores. The expected view, and there are some aspects that DTBQ results suggested that DTBQ validly measured does not cover. For example, side effects other diabetic treatment burden. Moreover, the than hypoglycemia (e.g., weight gain, gas- results of pairwise comparisons of DTBQ/DTSQ trointestinal symptoms) and financial aspects scores among types of diabetic treatment (e.g., cost of drug and device) are not included. (Table S4 in the electronic supplementary Other questionnaires measuring the uncovered material) showed that the mean DTBQ total aspects should be used together depending on burden score and DTBQ satisfaction score were the purpose, such as OHA-Q [36]. The original inversely related to the types of diabetic treat- version of DTBQ was written in Japanese; we ment. This indicated that the heavier burden of translated the questionnaire into English for diabetic treatment might worsen treatment this publication. If there are plans to use the satisfaction. English version with patients, a validation study In the studies comparing QW and BID for- will be needed. mulations of GLP-1 receptor agonist, better It has been already shown that improving treatment satisfaction was observed in patients treatment satisfaction and quality of life of treated with QW medication [24, 25]. Looking patients has a positive impact on their adher- at another therapeutic area, some observa- ence and BG control [37, 38]. In this study, tional/database studies on pharmacotherapy of DTBQ total burden score correlated negatively osteoporosis concluded that QW dosing was with DTSQ satisfaction score. Combining these associated with better adherence and persis- results, it is expected that increase of the tence than QD dosing [26–28]. These results treatment burden leads to worsening BG con- were consistent with the differences among trol. In ‘‘Standards of Medical Care in Dia- type of diabetic treatment observed in the betes—2017’’ released by the American DTBQ scores. Diabetes Association [39], the following state- Recently, patient preference of diabetic ments were included: treatment has been studied and reported • Treatment decision should be timely, rely on intensively [29–35]. For OHAs, a daily dosing evidence-based guidelines, and be made col- laboratively with patients based on individ- schedule was an important factor in addition to efficacy, safety, and cost. For injections, dosing ual preferences, prognosis, and comorbidities. frequency was the factor which most affected patient preference among all characteristics of • Providers should consider the burden of GLP-1 receptor agonists, including type of treatment and self-efficacy of patients when delivery system, nausea, hypoglycemia, BG recommending treatment. change, and dosing frequency where QW was 1016 Diabetes Ther (2018) 9:1001–1019 These statements reflect the importance of Disclosures. Hitoshi Ishii has received patient preference and treatment burden for research grants from Eli Lilly Japan and hono- determining the most appropriate treatment raria from Ono Pharmaceutical, Shionogi, Mit- option for diabetes. DTBQ is very much in the subishi Tanabe Pharma, Arkray, MSD, Taisho same direction indicated by these statements, Toyama Pharmaceutical, Becton–Dickinson and it enables quantitative measurement of the Japan, Novo Nordisk Pharma, Daiichi Sankyo, treatment burden caused by pharmacotherapy. Takeda Pharmaceutical, Astellas Pharma, and Measuring the treatment burden on patients Boehringer Ingelheim Japan, and lecture and/or with poor adherence would help to better consultant fees from Takeda Pharmaceutical, Eli understand their reasons, and some counter- Lilly Japan, Sanofi, MSD, Astellas Pharma, measures might be taken to improve their Novartis Pharma, Mitsubishi Tanabe Pharma, adherence. Daiichi Sankyo, Ono Pharmaceutical, AstraZe- In conclusion, on the basis of the above neca, Taisho Toyama Pharmaceutical, Shionogi, results and the considerations discussed, DTBQ Kowa Pharmaceutical, Boehringer Ingelheim has adequate reproducibility and validity as a Japan, Novo Nordisk Pharma, Sumitomo measurement scale for treatment burden on Dainippon Pharma, Kyowa Hakko Kirin, Ter- T2DM patients. umo Co., Bristol-Myers Squibb Co., Sanwa Kagaku Kenkyusho Co. Ltd., Fujirebio Inc., Kowa Pharmaceutical, Pfizer, and Teijin Pharma. Hiroki Shin has no conflict of interests. ACKNOWLEDGEMENTS Takahiro Tosaki has received lecture fees from Eli Lilly Japan, Takeda Pharmaceutical, and MSD. Tatsuya Haga has received lecture fees Funding. This study was funded by Eli Lilly from Eli Lilly Japan and Novo Nordisk Pharma. Japan (Kobe, Japan). No funding was received Yoshiki Nakajima has no conflict of interests. for the article processing charges and medical Toshihiko Shiraiwa has received research grants writing/editorial assistance. from Novo Nordisk Pharma, Sanofi, Takeda Pharmaceutical, AstraZeneca, Boehringer Ingel- Authorship. All named authors meet the heim Japan, and Quintiles and lecture fees from International Committee of Medical Journal Takeda Pharmaceutical and Sanofi. Nobuaki Editors (ICMJE) criteria for authorship for this Watanabe has received research grants from article, take responsibility for the integrity of Novo Nordisk Pharma and Astellas Pharma. the work as a whole, and have given their Miyuki Koizumi has received lecture fees from approval for this version to be published. Takeda Pharmaceutical. Hiroki Nakajima has Hitoshi Ishii was responsible for study design, received lecture fees from Astellas Pharma and interpretation of the results, and drafting this Takeda Pharmaceutical. Sadanori Okada has manuscript. All authors contributed to the data received lecture fees from MSD, Eli Lilly Japan, collection, reviewing, and final approval of this Takeda Pharmaceutical, Ono Pharmaceutical, manuscript. All authors had full access to all of Novo Nordisk Pharma, AstraZeneca, Mitsubishi the data in this study and take complete Tanabe Pharma, Sumitomo Dainippon Pharma, responsibility for the integrity of the data and Boehringer Ingelheim Japan, Sanofi, and Ark- accuracy of the data analysis. ray. Tsuyoshi Mashitani has received lecture Medical Writing and Editorial Assis- fees from MSD, Eli Lilly Japan, Boehringer tance. The authors thank Eisei Oda and Junko Ingelheim Japan, Kissei Pharmaceutical, Takeda Yamada in Medical TOUKEI Corp. (Tokyo, Pharmaceutical, Ono Pharmaceutical, Novo Japan) for data management, statistical analysis, Nordisk Pharma, Daiichi Sankyo, Mitsubishi and medical writing. No funding was received Tanabe Pharma, Sanofi, and AstraZeneca. for this. Takako Mohri has received lecture fees from Novo Nordisk Pharma, Takeda Pharmaceutical, Boehringer Ingelheim Japan, and Eli Lilly Japan. Diabetes Ther (2018) 9:1001–1019 1017 US diabetes care, 1999–2010. N Engl J Med. Yasuhiro Akai has received lecture fees from 2013;368(17):1613–24. MSD, Astellas Pharma, Otsuka Pharmaceutical, Ono Pharmaceutical, Kinden, Kissei Pharma- 3. Cramer JA. A systematic review of adherence with ceutical, Sanofi, Daiichi Sankyo, Sumitomo medications for diabetes. Diabetes Care. 2004;27(5):1218–24. Dainippon Pharma, Mitsubishi Tanabe Pharma, Chugai Pharmaceutical, Teijin Pharma, Torii 4. Donnelly LA, Morris AD, Evans JM, DARTS/MEMO Pharmaceutical, Eli Lilly Japan, Boehringer collaboration. Adherence to insulin and its associ- Ingelheim Japan, Novo Nordisk Pharma, Baxter, ation with glycaemic control in patients with type 2 diabetes. QJM. 2007;100(6):345–50. and Bristol-Myers Squibb. 5. Farmer AJ, Rodgers LR, Lonergan M, et al. Adher- Compliance with Ethics Guidelines. A ence to oral glucose-lowering therapies and associ- common protocol was approved at institutional ations with 1-year HbA1c: a retrospective cohort review boards in Nara Medical University and analysis in a large primary care database. Diabetes Care. 2016;39(2):258–63. TDE Healthcare Corporation TOSAKI Clinic for Diabetes and Endocrinology. All procedures 6. Bailey CJ, Kodack M. Patient adherence to medica- performed in studies involving human partici- tion requirements for therapy of type 2 diabetes. Int pants were in accordance with the ethical J Clin Pract. 2011;65(3):314–22. standards of the institutional and/or national ´ ´ ´ 7. Garcıa-Perez LE, Alvarez M, Dilla T, Gil-GuillenV, research committee and with the 1964 Decla- Orozco-Beltran D. Adherence to therapies in ration of Helsinki and its later amendments or patients with type 2 diabetes. Diabetes Ther. comparable ethical standards. Informed con- 2013;4(2):175–94. sent was obtained from all individual partici- 8. Kirkman MS, Rowan-Martin MT, Levin R, et al. pants included in the study. Determinants of adherence to diabetes medica- tions: findings from a large pharmacy claims data- Data Availability. The datasets generated base. Diabetes Care. 2015;38(4):604–9. and/or analyzed during the current study are 9. Polonsky WH, Henry RR. Poor medication adher- available from the corresponding author on ence in type 2 diabetes: recognizing the scope of the reasonable request. problem and its key contributors. Patient Prefer Adherence. 2016;10:1299–307. Open Access. This article is distributed 10. Peron EP, Ogbonna KC, Donohoe KL. Antidiabetic under the terms of the Creative Commons medications and polypharmacy. Clin Geriatr Med. Attribution-NonCommercial 4.0 International 2015;31(1):17–27. License (http://creativecommons.org/licenses/ by-nc/4.0/), which permits any non- 11. Schernthaner G. Fixed-dose combination therapies in the management of hyperglycaemia in type 2 commercial use, distribution, and reproduction diabetes: an opportunity to improve adherence and in any medium, provided you give appropriate patient care. Diabet Med. 2010;27(7):739–43. credit to the original author(s) and the source, provide a link to the Creative Commons license, 12. Blonde L, San Juan ZT. Fixed-dose combinations for treatment of type 2 diabetes mellitus. Adv Ther. and indicate if changes were made. 2012;29(1):1–13. 13. Polonsky WH, Fisher L, Hessler D, Bruhn D, Best JH. Patient perspectives on once-weekly medications REFERENCES for diabetes. Diabetes Obes Metab. 2011;13(2):144–9. 1. Thrasher J. Pharmacologic management of type 2 14. Kaku K. First novel once-weekly DPP-4 inhibitor, diabetes mellitus: available therapies. Am J Cardiol. trelagliptin, for the treatment of type 2 diabetes 2017;120(1S):S4–16. mellitus. Expert Opin Pharmacother. 2015;16(16):2539–47. 2. Ali MK, Bullard KM, Saaddine JB, Cowie CC, Imperatore G, Gregg EW. Achievement of goals in 1018 Diabetes Ther (2018) 9:1001–1019 15. Evans PM, Bain SC. Omarigliptin for the treatment 27. Kishimoto H, Maehara M. Compliance and persis- of type 2 diabetes mellitus. Expert Opin Pharma- tence with daily, weekly, and monthly bisphos- cother. 2016;17(14):1947–52. phonates for osteoporosis in Japan: analysis of data from the CISA. Arch Osteoporos. 2015;10:231. 16. Brunton S, Davidson JA. Exenatide once weekly: a review of pharmacology and treatment considera- 28. Iglay K, Cao X, Mavros P, Joshi K, Yu S, Tunceli K. tions in type 2 diabetes. Clin Ther. Systematic literature review and meta-analysis of 2016;38(3):582–94. medication adherence with once-weekly versus once-daily therapy. Clin Ther. 2015;37(8):1813–21. 17. Burness CB, Scott LJ. Dulaglutide: a review in type 2 diabetes. BioDrugs. 2015;29(6):407–18. 29. Hauber AB, Han S, Yang JC, et al. Effect of pill burden on dosing preferences, willingness to pay, 18. Qiao Q, Ouwens MJ, Grandy S, Johnsson K, Kostev and likely adherence among patients with type 2 K. Adherence to GLP-1 receptor agonist therapy diabetes. Patient Prefer Adherence. 2013;7:937–49. administered by once-daily or once-weekly injec- tion in patients with type 2 diabetes in Germany. 30. Hauber AB, Tunceli K, Yang JC, et al. A survey of Diabetes Metab Syndr Obes. 2016;9:201–5. patient preferences for oral antihyperglycemic therapy in patients with type 2 diabetes mellitus. Diabetes Ther. 2015;6(1):75–84. 19. Nguyen H, Dufour R, Caldwell-Tarr A. Glucagon- like peptide-1 receptor agonist (GLP-1RA) therapy adherence for patients with type 2 diabetes in a 31. Gelhorn HL, Poon JL, Davies EW, Paczkowski R, medicare population. Adv Ther. 2017;34(3):658–73. Curtis SE, Boye KS. Evaluating preferences for pro- files of GLP-1 receptor agonists among injection- 20. Polonsky WH, Anderson BJ, Lohrer PA, et al. naı¨ve type 2 diabetes patients in the UK. Patient Assessment of diabetes-related distress. Diabetes Prefer Adherence. 2015;9:1611–22. Care. 1995;18(6):754–60. 32. Hauber AB, Nguyen H, Posner J, Kalsekar I, Ruggles 21. Polonsky WH, Fisher L, Earles J, et al. Assessing J. A discrete-choice experiment to quantify patient psychosocial distress in diabetes: development of preferences for frequency of glucagon-like peptide- the diabetes distress scale. Diabetes Care. 1 receptor agonist injections in the treatment of 2005;28(3):626–31. type 2 diabetes. Curr Med Res Opin. 2016;32(2):251–62. 22. Dennick K, Sturt J, Speight J. What is diabetes dis- tress and how can we measure it? A narrative review 33. Gelhorn HL, Bacci ED, Poon JL, Boye KS, Suzuki S, and conceptual model. J Diabetes Complicat. Babineaux SM. Evaluating preferences for profiles of 2017;31(5):898–911. glucagon-like peptide-1 receptor agonists among injection-naive type 2 diabetes patients in Japan. 23. Pouwer F, Snoek FJ, van der Ploeg HM, Heine RJ, Patient Prefer Adherence. 2016;10:1337–48. Brand AN. A comparison of the standard and the computerized versions of the Well-being Ques- 34. Qin L, Chen S, Flood E, et al. Glucagon-like peptide- tionnaire (WBQ) and the Diabetes Treatment Sat- 1 receptor agonist treatment attributes important to isfaction Questionnaire (DTSQ). Qual Life Res. injection-experienced patients with type 2 diabetes 1998;7:33–8. mellitus: a preference study in Germany and the United Kingdom. Diabetes Ther. 2017;8(2):335–53. 24. Best JH, Boye KS, Rubin RR, Cao D, Kim TH, Peyrot M. Improved treatment satisfaction and weight-re- 35. Qin L, Chen S, Flood E, et al. Glucagon-like peptide- lated quality of life with exenatide once weekly or 1 receptor agonist treatment attributes important to twice daily. Diabet Med. 2009;26(7):722–8. injection-naı¨ve patients with type 2 diabetes mel- litus: a multinational preference study. Diabetes 25. Reaney M, Yu M, Lakshmanan M, Pechtner V, van Ther. 2017;8(2):321–34. Brunt K. Treatment satisfaction in people with type 2 diabetes mellitus treated with once-weekly 36. Ishii H, Oda E. Reproducibility and validity of a dulaglutide: data from the AWARD-1 and AWARD- satisfaction questionnaire on hypoglycemic agents: 3 clinical trials. Diabetes Obes Metab. the Oral Hypoglycemic Agent Questionnaire (OHA- 2015;17(9):896–903. Q). Diabetol Int. 2012;3(3):152–63. 26. Confavreux CB, Canoui-Poitrine F, Schott AM, 37. Roborel de Climens A, Tunceli K, Arnould B, et al. Ambrosi V, Tainturier V, Chapurlat RD. Persistence Review of patient-reported outcome instruments measuring health-related quality of life and satis- at 1 year of oral antiosteoporotic drugs: a prospec- faction in patients with type 2 diabetes treated with tive study in a comprehensive health insurance database. Eur J Endocrinol. 2012;166(4):735–41. oral therapy. Curr Med Res Opin. 2015;31(4):643–65. Diabetes Ther (2018) 9:1001–1019 1019 38. Ishii H, Anderson JH Jr, Yamamura A, Takeuchi M, 39. American Diabetes Association. Promoting health Ikeda I. Improvement of glycemic control and and reducing disparities in populations. Diabetes quality-of-life by insulin lispro therapy: assessing Care. 2017;40(Suppl 1):S6–10. benefits by ITR-QOL questionnaires. Diabetes Res Clin Pract. 2008;81(2):169–78. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Diabetes Therapy Springer Journals

Reproducibility and Validity of a Questionnaire Measuring Treatment Burden on Patients with Type 2 Diabetes: Diabetic Treatment Burden Questionnaire (DTBQ)

Free
19 pages

Loading next page...
 
/lp/springer_journal/reproducibility-and-validity-of-a-questionnaire-measuring-treatment-f1jNrGeuyZ
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Medicine & Public Health; Internal Medicine; Diabetes; Cardiology; Endocrinology
ISSN
1869-6953
eISSN
1869-6961
D.O.I.
10.1007/s13300-018-0414-4
Publisher site
See Article on Publisher Site

Abstract

Diabetes Ther (2018) 9:1001–1019 https://doi.org/10.1007/s13300-018-0414-4 ORIGINAL RESEARCH Reproducibility and Validity of a Questionnaire Measuring Treatment Burden on Patients with Type 2 Diabetes: Diabetic Treatment Burden Questionnaire (DTBQ) . . . . Hitoshi Ishii Hiroki Shin Takahiro Tosaki Tatsuya Haga . . . Yoshiki Nakajima Toshihiko Shiraiwa Nobuaki Watanabe . . . . Miyuki Koizumi Hiroki Nakajima Sadanori Okada Tsuyoshi Mashitani Takako Mohri Yasuhiro Akai Received: January 16, 2018 / Published online: March 29, 2018 The Author(s) 2018 Treatment Burden Questionnaire (DTBQ), a ABSTRACT patient-administered questionnaire composed of 18 questions, and evaluated its reproducibil- Introduction: To measure the burden of phar- ity and validity. macotherapy on patients with type 2 diabetes Methods: We enrolled 240 patients with T2DM mellitus (T2DM), we developed the Diabetes under pharmacotherapy over 20 years of age at seven institutes in Japan. Their physicians filled Enhanced content To view enhanced content for this out report forms on patient backgrounds, and article go to https://doi.org/10.6084/m9.figshare. the patients answered both the DTBQ and the Diabetes Treatment Satisfaction Questionnaire Electronic supplementary material The online (DTSQ). For evaluation of reproducibility, 48 of version of this article (https://doi.org/10.1007/s13300- 018-0414-4) contains supplementary material, which is the enrolled subjects completed a 2nd DTBQ at available to authorized users. home after leaving the medical institutes. Results: Statistical analyses were performed for H. Ishii (&)  M. Koizumi  H. Nakajima  S. Okada two sets of subjects, the validity analysis set T. Mashitani  T. Mohri  Y. Akai (N = 236) and the reproducibility analysis set Department of Diabetology, Nara Medical (N = 47). Factor analysis found a simple struc- University, Nara, Japan e-mail: hits1@naramed-u.ac.jp ture in the DTBQ item scores using a three-fac- tor model with varimax rotation; the three H. Shin subscales were designated as ‘‘implementation Shin Clinic, Tokyo, Japan burden’’, ‘‘flexibility burden’’, and ‘‘blood glu- T. Tosaki cose control burden’’. All intraclass correlation TDE Healthcare Corporation TOSAKI Clinic for coefficients for the subscale scores were 0.8 or Diabetes and Endocrinology, Aichi, Japan higher, indicating high reproducibility. Nega- T. Haga tive correlations were observed between the Haga Diabetes Clinic, Aichi, Japan DTSQ satisfaction score and the DTBQ subscale scores. Moreover, as the dosing frequency of Y. Nakajima Nakajima Diabetes Clinic, Osaka, Japan diabetic medicines increased, the DTBQ total score (total burden score) also became higher. T. Shiraiwa Likewise, expected associations were observed Shiraiwa Medical Clinic, Osaka, Japan between patient backgrounds and DTSQ scores. N. Watanabe Watanabe Medical Clinic, Hyogo, Japan 1002 Diabetes Ther (2018) 9:1001–1019 Conclusion: The DTBQ has adequate repro- therefore have the greater risk associated with ducibility and validity as a measurement scale polypharmacy [10]. Considering the factors that for treatment burden on T2DM patients. might be causes of non-adherence, a less com- Trial Registration: University Hospital Medical plex and more convenient pharmacotherapy Information Network (UMIN) 000026382. would be preferable, and this together with a Funding: Eli Lilly Japan. reduction of treatment burden is a key for improving treatment. One possible solution for reducing treatment Keywords: Pharmacotherapy; Questionnaire; burden is the use of single-tablet fixed-dose Reproducibility; Treatment burden; Type 2 combinations of multiple OHAs [11, 12]. Fixed- diabetes mellitus; Validity dose combination therapies require fewer tablets to be taken and lessen treatment com- INTRODUCTION plexity, and thereby reduce treatment burden by replacing multiple agents with one fixed- Pharmacotherapy for type 2 diabetes mellitus dose combination. Another solution is the use (T2DM) has made remarkable progress since the of a long-acting formulation with a hypo- glycemic effect sustained for 1 week [13]. Cur- introduction of insulin, the first injectable treatment for diabetes, in the 1920s, rently, there are two available once-weekly (QW) formulations: an OHA DPP-4 inhibitor and sulfonylureas, the first oral hypoglycemic agent (OHA), in the 1950s [1]. Though the pro- and an injectable GLP-1 receptor agonist gress of pharmacotherapy has contributed to [14–17]. Some observational studies reported improving blood glucose (BG) control in T2DM that T2DM patients treated with QW GLP-1 patients, a substantial number of patients are receptor agonist exhibited better adherence still poorly controlled. In a survey of T2DM than patients treated with a once-daily (QD) adults in the USA, only 52.2% of the subjects agent [18, 19]. QW agents are largely equivalent achieved a glycemic target of HbA1c \ 7.0% in to the corresponding QD agents in terms of 2007–2010 [2]. Increasing the number of T2DM efficacy (i.e., hypoglycemic effect), safety (i.e., adverse events including hypoglycemia), and patients with a sufficient level of BG control is an important medical objective, and one way of cost (i.e., drug price). Therefore, if switching from QD to QW agents improved adherence, achieving this is to improve patient adherence to treatment. This is supported by a report which the reason would be that the treatment burden showed that a considerable portion of patients was reduced by mitigating the complexity of exhibited poor adherence [3] and some obser- QD treatment. However, there is currently no vational studies that also showed a positive questionnaire available for measuring quanti- correlation between adherence to medication tatively the difference of treatment burden and BG control level in the real world [4, 5]. between QW and QD agents. To measure burden on patients with dia- Multiple factors are considered to affect non- adherence to diabetic treatment [6–9]. Patient betes, two scales, PAID (Problem Areas in Dia- betes Survey) and DDS (Diabetes Distress Scale) factors include demographics (e.g., age, sex), psychological factors (e.g., health belief), phys- were developed so far [20–22]. However, these scales focus on the generic burden ical factors (e.g., concomitant disease), and social and economic status (e.g., education, attributable to diabetes, not on the treatment income). On the other hand, there are specific burden caused by pharmacotherapy. Therefore, pharmacotherapy factors: efficacy, safety, cost, they are not suitable for differentiating medici- complexity (e.g., frequency and interval of nes according to their characteristics, such as dose), and convenience (e.g., ease of swallowing dosing schedule which dominantly determines formulation, ease of handling injection device). the complexity and convenience of implemen- tation of pharmacotherapy. Thus, to measure In addition, T2DM patients, especially older adults, commonly have concomitant diseases, the treatment burden caused by pharma- cotherapy for T2DM, we developed DTBQ such as hypertension and dyslipidemia, and Diabetes Ther (2018) 9:1001–1019 1003 (Diabetes Treatment Burden Questionnaire) • Injection of GLP-1 receptor agonist, once weekly ± OHA (Injection/QW) focusing on efficacy (i.e., BG control), safety (i.e., hypoglycemia), complexity (i.e., amount, • Injection of insulin or GLP-1 receptor ago- nist, once daily ± OHA (Injection/QD) frequency, timing, and interval of dose), and convenience (i.e., ease of taking or injecting • Injection of insulin or GLP-1 receptor ago- medicine), and evaluated its reproducibility and nist, twice a day or more ± OHA (Injection/ validity. This questionnaire would make it pos- BID?) sible to quantify the treatment burden on • Only OHA, once weekly (OHA/QW) T2DM patients depending on the characteristics • Only OHA, once daily (OHA/QD) of medicines. • Only OHA, twice a day or more (OHA/BID?) In addition, subjects had to be at least 20 years old and provide written informed METHODS consent. Moreover, the following patients were excluded: patients who were diagnosed with or Development of DTBQ suspected to have dementia, patients with psy- chiatric disorders, patients without sufficient We interviewed 16 patients with T2DM treated judgment ability, patients who needed consent with pharmacotherapy about treatment burden. from their proxies, and other patients judged to Based on the interview, pre-existing question- be inappropriate by the attending physicians. naires, and opinions from diabetologists, the first The target numbers of enrolled patients were version of questionnaire with 26 questions was the following: 210 subjects for validity evalua- drafted. This first version was completed by 20 tion who answered DTBQ at least once, and 70 patients for pilot testing. Based on these results, subjects for reproducibility evaluation who 18 questions were selected for the final version. answered DTBQ twice. The enrollment was Each question is answered by choosing one of adjusted to balance the number of patients seven ordered categories: (1) strongly disagree, among the different types of diabetic treatment. (2) mostly disagree, (3) slightly disagree, (4) Patients enrolled as diabetic treatment includ- neither agree nor disagree, (5) slightly agree, (6) ing injectable medicines had the option to use mostly agree, (7) strongly agree. It should be OHAs together, so they were asked to answer noted that the relations between the category the questions in DTBQ on the number and level of treatment burden are injectable medicines only. opposite for questions 1–10 and questions The attending physician gave the ‘‘Report 11–18. In other words, a larger category number form for all patients’’ along with the 1st DTBQ means heavier treatment burden in questions and DTSQ to be filled out to the patient after 1–10, but a smaller burden in questions 11–18. confirming his/her eligibility; the ‘‘Report form Therefore, item scores are defined by the cate- for reproducibility evaluation’’, which included gory number converted so that 0 means ‘‘mini- the 2nd DTBQ, was then delivered only if the mum treatment burden’’ and 6 means patient was a subject for reproducibility evalu- ‘‘maximum treatment burden’’. Correspondence ation. The patient filled out the ‘‘Report form between the category number and the item score for all patients’’ without the presence of the for each question of DTBQ is shown in Table S1 attending physician in a room at the medical in the electronic supplementary material. institute. Then the patient put the form in an envelope and posted it to the data center or gave it to the attending physician. The attend- Evaluation of Reproducibility and Validity ing physician filled out the ‘‘Report form for physicians’’ and posted it to the data center All subjects were outpatients with T2DM treated with the ‘‘Report form for all patients’’ if he had for at least 12 weeks with pharmacotherapy received it from the patient. Patients who were corresponding to one of the following six types subjects for reproducibility evaluation filled out of diabetic treatment: the ‘‘Report form for reproducibility evaluation’’ 1004 Diabetes Ther (2018) 9:1001–1019 and posted it to the data center after leaving the for the validity analysis set. Based on the results, medical institute. subscales were determined and named. Intra- DTSQ is the most popular patient-reported class correlation coefficients (ICC) were calcu- questionnaire to evaluate patient satisfaction in lated for the DTBQ subscale and total scores for diabetic treatment [23]. License and copyright the reproducibility analysis set. Correlation agreement for using DTSQ questionnaire in this coefficients of the DTBQ subscale and total study was obtained from Health Psychology scores and the DTSQ scores were calculated for Research Limited (University of London, UK). the validity analysis set. In addition, analysis of The study was conducted from November 2016 variance (ANOVA) or the t test was used to to February 2017 at seven sites in Japan and was examine associations between DTBQ subscale registered at University Hospital Medical Infor- and total scores and patient background factors. mation Network (UMIN000026382). A common protocol was approved at insti- RESULTS tutional review boards in Nara Medical Univer- sity and TDE Healthcare Corporation TOSAKI Patients Clinic for Diabetes and Endocrinology. All pro- cedures performed in studies involving human Patient disposition is shown in Fig. 1. Out of participants were in accordance with the ethical standards of the institutional and/or national 240 enrolled patients, 236 were included in the validity analysis set; four were excluded because research committee and with the 1964 Decla- ration of Helsinki and its later amendments or of unanswered questions in the 1st DTBQ. The numbers of patients by type of diabetic treat- comparable ethical standards. Informed con- sent was obtained from all individual partici- ment were as follows: (1) Injection/QW, 40; (2) Injection/QD, 41; (3) Injection/BID?, 37; (4) pants included in the study. OHA/QW, 38; (5) OHA/QD, 35; and (6) OHA/ BID?, 45. Out of 236 patients in the validity Statistical Analyses analysis set, 48 patients filled out the 2nd DTBQ. One patient was excluded because of Enrolled patients who left any unanswered unanswered questions, and 47 were included questions in the 1st DTBQ were excluded from the reproducibility analysis set. the validity analysis set. Of the validity analysis Patient backgrounds reported by physicians set, patients who completed the 2nd DTBQ and patients are summarized in Table 1 and collected and without any unanswered ques- Table S2 (in the electronic supplementary tions were included in the reproducibility material), respectively. All background factors analysis set. Appropriate descriptive statistics were similarly distributed in the validity and were calculated for background factors, such as reproducibility analysis sets. type of diabetic treatment, HbA1c, age, and sex, for the validity and reproducibility analysis sets. Summary and Reproducibility of DTBQ Mean and standard deviation were calculated Item Scores for DTBQ item scores, and for the validity and reproducibility analysis sets. In addition, weighted kappa coefficients were calculated for Descriptive statistics of the DTBQ item scores for the validity and reproducibility analysis sets, the reproducibility analysis set where the weights (w) were assigned as follows: exact and weighted kappa coefficients for the repro- agreement, w = 1; one-level disagreement, ducibility analysis set are shown in Table 2. For all questions, item scores of 0 to 2, which indi- w = 5/6; two-level disagreement, w = 4/6; three- level disagreement, w = 3/6; four-level dis- cate a relatively small burden, were reported from more than half of the patients in the agreement, w = 2/6; five-level disagreement, w = 1/6; six-level disagreement, w = 0/6. Factor validity analysis set. The averages of 14 item scores out of the 18 questions were less than 2.0. analysis of the DTBQ item scores was conducted Diabetes Ther (2018) 9:1001–1019 1005 Fig. 1 Patient disposition. Full description of type of agent, once a week; 5. Oral hypoglycemic agent, once a day; diabetic treatment: 1. Injection of GLP-1 agonist, once a 6. Oral hypoglycemic agent, twice a day or more. BID? week ± oral hypoglycemic agent; 2. Injection of insulin or twice a day or more, DTBQ Diabetic Treatment Burden GLP-1 agonist, once a day ± oral hypoglycemic agent; 3. Questionnaire, OHA oral hypoglycemic agent, QD once Injection of insulin or GLP-1 agonist, twice a day or daily, QW once weekly more ± oral hypoglycemic agent; 4. Oral hypoglycemic Four questions whose average scores were medication (j = 0.550)’’, ‘‘8. Concern about greater than 2.0 were as follows: ‘‘7. Medication hypoglycemia (j = 0.532)’’, ‘‘17. Feeling less away from home’’, ‘‘13. Allowing me to take a burden to continue diabetes treatment missed dose’’, ‘‘16. Feeling less burden to follow (j = 0.560)’’, and ‘‘18. Satisfaction with my diet therapy’’, and ‘‘18. Satisfaction with my current BG control (j = 0.579)’’, which indi- current BG control’’. cated reproducibility good enough to use in Means of the 1st and 2nd item scores in the practice. reproducibility analysis set were close to those in the validity analysis set. Weighted kappa Factor Analysis of DTBQ Item Scores coefficients for the reproducibility analysis set were estimated at 0.6 or higher for 14 of the 18 Table 3 shows the results of factor analysis of questions. Four questions with point estimates the DTBQ item scores for the validity analysis less than 0.6 were ‘‘3. Securing time for 1006 Diabetes Ther (2018) 9:1001–1019 Table 1 Patient backgrounds reported by physicians Background factors Validity analysis set Reproducibility analysis set Number of patients 236 (100.0%) 47 (100.0%) Type of diabetic treatment Injection/QW 40 (16.9%) 10 (21.3%) Injection/QD 41 (17.4%) 8 (17.0%) Injection/BID+ 37 (15.7%) 5 (10.6%) OHA/QW 38 (16.1%) 7 (14.9%) OHA/QD 35 (14.8%) 5 (10.6%) OHA/BID+ 45 (19.1%) 12 (25.5%) HbA1c (%), mean (SD) 6.98 (0.92) 7.28 (1.14) Age (years), mean (SD) 63.4 (11.9) 62.8 (12.0) Sex Male 142 (60.2%) 27 (57.4%) Female 94 (39.8%) 20 (42.6%) Duration of diabetes (years), mean (SD) 13.1 (9.6) 13.5 (12.4) Frequency of dose for injection None 118 (50.0%) 24 (51.1%) Once a week 40 (16.9%) 10 (21.3%) Once a day 41 (17.4%) 8 (17.0%) Twice a day or more 37 (15.7%) 5 (10.6%) Type of injection Insulin 68 (28.8%) 13 (27.7%) GLP-1 agonist (QD, BID) 12 (5.1%) 0 (0.0%) GLP-1 agonist (QW) 40 (16.9%) 10 (21.3%) Frequency of dose for OHA None 21 (8.9%) 3 (6.4%) Once a week 40 (16.9%) 7 (14.9%) Once a day 62 (26.3%) 11 (23.4%) Twice a day 48 (20.3%) 9 (19.1%) Three times a day or more 65 (27.5%) 17 (36.2%) Type of OHA Sulfonylureas 61 (25.8%) 15 (31.9%) Glinide 39 (16.5%) 6 (12.8%) DPP-4 inhibitor 128 (54.2%) 25 (53.2%) Diabetes Ther (2018) 9:1001–1019 1007 Table 1 continued Background factors Validity analysis set Reproducibility analysis set Thiazolidine derivative 24 (10.2%) 4 (8.5%) Biguanide 81 (34.3%) 18 (38.3%) a-Glucosidase inhibitor 45 (19.1%) 14 (29.8%) SGLT2 inhibitor 30 (12.7%) 8 (17.0%) Diabetic neuropathy No 147 (62.3%) 27 (57.4%) Yes 52 (22.0%) 15 (31.9%) Unknown 37 (15.7%) 5 (10.6%) Diabetic nephropathy No 174 (73.7%) 30 (63.8%) Yes 62 (26.3%) 17 (36.2%) Diabetic retinopathy No 150 (63.6%) 30 (63.8%) Yes 61 (25.8%) 16 (34.0%) Unknown 25 (10.6%) 1 (2.1%) BID+ twice a day or more, OHA oral hypoglycemic agent, QD once daily, QW once weekly set. In the factor pattern for a four-factor model Reproducibility of DTBQ Subscale Scores without rotation, the first factor loadings were and Total Burden Score more than 0.4 for all questions, which indicated a unidimensional feature of DTBQ. Therefore, Scatter plots of the DTBQ subscale scores and the total score of all 18 questions was desig- total burden score (1st vs. 2nd) and the esti- nated as the ‘‘total burden score’’. On the other mates of ICC for the reproducibility analysis set hand, the variance of three factors was greater are shown in Fig. 2. The ICC estimates were 0.8 than 1.0 in the four-factor model. or higher, indicating high reproducibility of all In the factor pattern for three-factor model scores. with varimax rotation, focusing on factor load- ing more than 0.4, a simple structure was Correlation Between DTBQ and DTSQ obtained, where all questions belonged to one Scores of the three factors. Considering the factor pattern and contents of the questions, three Table 4 shows the univariate descriptive statis- subscales were determined and designated: tics and correlations of the DTBQ and DTSQ • Implementation burden score: sum of item scores for the validity analysis set. Correlations scores 1–10 between the DTBQ subscale scores were 0.4 or • Flexibility burden score: sum of item scores higher for Spearman’s correlation coefficient 11–13 (SCC) for every combination. The DTBQ total • Blood control (BG) control burden score: burden scores correlated positively with all the sum of item scores 14–18 1008 Diabetes Ther (2018) 9:1001–1019 Table 2 Reproducibility of DTBQ item scores (reproducibility analysis set) Questions abbreviated Reproducibility analysis set (N = 47) 1st mean 2nd mean Weighted j coefficient (SD) (SD) Point 95% confidence estimate interval 1. Medication on time 1.7 (1.6) 1.5 (1.6) 0.785 (0.665, 0.906) 2. Medication during busy hours 1.7 (1.7) 1.6 (1.6) 0.807 (0.710, 0.904) 3. Securing time for medication 0.9 (1.1) 1.1 (1.2) 0.550 (0.355, 0.744) 4. Pain associated with medication 1.3 (1.8) 1.3 (1.7) 0.690 (0.538, 0.841) 5. Feeling that I should not miss a dose 1.8 (1.7) 1.5 (1.5) 0.697 (0.549, 0.846) 6. Feeling guilty when I miss a dose 1.7 (1.9) 1.6 (1.8) 0.839 (0.752, 0.926) 7. Medication away from home 1.9 (1.8) 2.0 (1.9) 0.799 (0.705, 0.892) 8. Concern about hypoglycemia 0.9 (1.2) 1.1 (1.2) 0.532 (0.333, 0.731) 9. Inflexibility to adjust the time for medication 1.0 (1.3) 1.2 (1.3) 0.677 (0.505, 0.848) 10. Worrying about future 1.8 (1.7) 1.9 (1.7) 0.710 (0.579, 0.840) 11. Short time and small effort for medication 1.1 (1.5) 1.1 (1.2) 0.627 (0.434, 0.821) 12. Medication without time pressure 2.0 (1.8) 1.9 (1.9) 0.811 (0.702, 0.920) 13. Allowing me to take a missed dose 1.9 (2.1) 1.9 (1.9) 0.791 (0.693, 0.889) 14. Feeling that my diabetes is getting better 1.9 (1.3) 2.0 (1.3) 0.713 (0.586, 0.839) 15. Allowing me to control BG with small effort 1.8 (1.3) 1.7 (1.3) 0.650 (0.484, 0.816) 16. Feeling less burden to follow diet therapy 2.0 (1.5) 2.1 (1.5) 0.643 (0.481, 0.804) 17. Feeling less burden to continue diabetes treatment 1.5 (1.4) 1.7 (1.3) 0.560 (0.392, 0.728) 18. Satisfaction with my current BG control 2.0 (1.5) 2.1 (1.7) 0.579 (0.414, 0.745) Range of item scores: 0 (minimum treatment burden)–6 (maximum treatment burden) DTBQ subscale scores with 0.7 or higher SCC Association Between DTBQ Scores estimates. and Patient Backgrounds The DTSQ hyperglycemic score correlated positively with the DTBQ scores: implementa- For the validity analysis set, analysis results of tion burden (SCC = 0.31), BG control burden the association between DTBQ scores and (0.36), and total burden (0.36). Weak correla- patient backgrounds reported by physicians and tions were observed between the DTSQ hypo- patients are shown in Table 5 and Table S3 (in glycemic score and the DTBQ scores within the the electronic supplementary material), respec- range of ± 0.2 SCC. The DTSQ satisfaction score tively. In addition, the results of pairwise com- correlated negatively with all the DTBQ scores parisons of the DTBQ/DTSQ scores between with - 0.3 SCC or less. In particular, the BG types of diabetic treatment are shown in control burden score (SCC = - 0.61) and total Table S4 in the electronic supplementary burden score (- 0.53) were correlated strongly. material. Diabetes Ther (2018) 9:1001–1019 1009 Table 3 Factor analysis of DTBQ item scores (validity analysis set) Questions abbreviated 4-factor model without rotation 3-factor model with varimax rotation Factor loading Factor loading Communality 12 3 4 12 3 1. Medication on time 0.743* - 0.321 - 0.004 - 0.124 0.755* 0.164 0.112 0.609 2. Medication during busy 0.734* - 0.393 0.003 0.006 0.817* 0.105 0.081 0.685 hours 3. Time ensure for medication 0.645* - 0.329 0.022 0.300 0.694* 0.095 0.097 0.500 4. Pain associated with 0.408* - 0.293 - 0.164 0.449* 0.479* 0.088 - 0.088 0.245 medication 5. To feel that I should not 0.705* - 0.384 - 0.036 - 0.137 0.803* 0.112 0.044 0.659 miss a dose 6. To feel guilty when I miss a 0.464* - 0.276 0.110 - 0.334 0.527* - 0.004 0.118 0.291 dose 7. Medication away from home 0.711* - 0.248 0.043 - 0.105 0.720* 0.162 0.169 0.573 8. Concern about 0.503* - 0.108 0.023 0.210 0.462* 0.163 0.144 0.261 hypoglycemia 9. Not to flexibly adjust the 0.664* - 0.242 0.086 - 0.092 0.675* 0.122 0.188 0.506 time for medication 10. To worry about future 0.604* - 0.190 - 0.093 - 0.048 0.591* 0.219 0.052 0.400 11. Short time and small effort 0.408* 0.331 0.326 0.260 0.124 0.245 0.530* 0.356 for medication 12. Medication without time 0.487* 0.426* 0.522* - 0.033 0.124 0.236 0.796* 0.704 pressure 13. To allow me to take a 0.427* 0.377 0.510* - 0.058 0.106 0.180 0.741* 0.592 missed dose 14. To feel that my diabetes is 0.422* 0.520* - 0.218 0.023 0.025 0.676* 0.189 0.493 getting better 15. To allow me to control BG 0.578* 0.528* - 0.243 - 0.043 0.144 0.780* 0.222 0.678 with small effort 16. To feel less burden to 0.616* 0.454* - 0.146 - 0.025 0.219 0.694* 0.284 0.611 follow diet therapy 17. To feel less burden to 0.611* 0.422* - 0.134 0.117 0.235 0.659* 0.280 0.568 continue diabetes treatment 18. Satisfied with my current 0.511* 0.391 - 0.423* - 0.165 0.180 0.730* 0.001 0.566 BG control Variance of factors 6.061 2.361 1.023 0.619 4.606 2.847 1.846 1010 Diabetes Ther (2018) 9:1001–1019 Table 3 continued Questions abbreviated 4-factor model without rotation 3-factor model with varimax rotation Factor loading Factor loading Communality 12 3 4 12 3 Cronbach’s a 0.885 0.887 0.862 0.775 N = 236. *Factor loading is more than 0.4 or less than - 0.4 Fig. 2 Reproducibility of DTBQ subscale scores and total 11–13), c BG control burden score (sum of scores for burden scores (reproducibility analysis set). a Implementa- items 14–18), d total burden score (sum of scores for items tion burden score (sum of scores for items 1–10), 1–18). CI confidence interval b flexibility burden score (sum of scores for items DTBQ implementation burden score had dose: injection and OHA, diabetic retinopathy) significant associations with three items repor- and three items reported by patients (i.e., ted by physicians (i.e., type of diabetic treat- hypoglycemic event, frequency of missed dose: ment, age, frequency of dose: OHA) and five injection or OHA, time to leave home on items reported by patients (i.e., BG control, weekdays). DTBQ flexibility burden score was hypoglycemic event, frequency of missed dose: high in the following patients: frequently dos- injection, OHA, injection or OHA). DTBQ ing with injection or OHA, higher HbA1c implementation burden scores were high in the (C 7.0%), older age (C 65 years), longer dura- following patients: treated with injection, fre- tion of diabetes (C 10 years), with diabetic quently dosing with injection or OHA, younger retinopathy, with hypoglycemic events, fre- age (\ 65 years), with BG controlled poorly, quently missing dose of injection or OHA, with hypoglycemic events, frequently missing leaving home on weekdays late (8:30 or later). dose of injection or OHA. DTBQ BG control burden score had signifi- DTBQ flexibility burden score had significant cant associations with four items reported by associations with seven items reported by physicians (i.e., type of diabetic treatment, physicians (i.e., type of diabetic treatment, HbA1c, diabetic nephropathy and retinopathy) HbA1c, age, duration of diabetes, frequency of and four items reported by patients (i.e., BG Diabetes Ther (2018) 9:1001–1019 1011 Table 4 Correlation between DTBQ scores and DTSQ scores (validity analysis set) Univariate statistics DTBQ DTSQ Implementation Flexibility BG control burden Total Hyperglycemia Hypoglycemia Satisfaction burden burden burden Mean 13.5 5.3 9.3 28.1 2.5 1.1 27.4 Standard deviation 11.6 4.9 6.0 17.5 1.8 1.5 6.1 N 236 236 236 236 236 236 236 Minimum 0 0 0 0 0 0 12 Median 10.0 5.0 9.0 27.0 2.5 0.0 28.5 Maximum 51 18 26 76 6 6 36 Correlation coefficient matrix: Spearman’s correlation coefficient DTBQ Implementation 1.00 – – – – – – burden Flexibility burden 0.43 1.00 – – – – – BG control burden 0.42 0.46 1.00 – – – – Total burden 0.86 0.70 0.73 1.00 – – – DTSQ Hyperglycemia 0.31 0.16 0.36 0.36 1.00 – – Hypoglycemia 0.14 0.10 0.03 0.12 0.25 1.00 – Satisfaction - 0.38 - 0.36 - 0.61 - 0.53 - 0.31 - 0.07 1.00 BG blood glucose, DTBQ Diabetic Treatment Burden Questionnaire, DTSQ Diabetic Treatment Satisfaction Questionnaire 1012 Diabetes Ther (2018) 9:1001–1019 Table 5 Associations between DTBQ scores and patient backgrounds reported by physicians (validity analysis set) Patients’ backgrounds N DTBQ scores Implementation burden Flexibility burden BG control burden Total burden Category Mean (SD) Mean (SD) Mean (SD) Mean (SD) Validity analysis set 236 13.5 (11.6) 5.3 (4.9) 9.3 (6.0) 28.1 (17.5) Type of diabetic treatment Injection/QW 40 12.3 (11.0) 2.7 (3.5) 9.4 (5.4) 24.3 (14.9) Injection/QD 41 18.3 (13.9) 6.6 (4.3) 9.3 (6.1) 34.2 (18.3) Injection/BID? 37 17.5 (13.2) 8.1 (5.4) 11.1 (6.7) 36.6 (20.4) OHA/QW 38 7.9 (8.3) 2.0 (2.7) 7.0 (4.3) 17.0 (12.0) OHA/QD 35 8.2 (7.0) 4.6 (4.5) 8.4 (6.8) 21.1 (12.9) OHA/BID? 45 15.8 (10.1) 7.6 (5.0) 10.6 (6.1) 33.9 (15.8) ANOVA P \ 0.001*** P \ 0.001*** P = 0.044* P \ 0.001*** HbA1c \ 7.0% 133 12.6 (11.6) 4.7 (4.5) 7.9 (5.3) 25.2 (16.6) 7.0% or higher 103 14.6 (11.5) 6.1 (5.3) 11.1 (6.5) 31.9 (17.9) t test P = 0.186 P = 0.022* P \ 0.001*** P = 0.004** Age \65 115 15.1 (11.5) 4.6 (4.1) 9.6 (5.9) 29.3 (17.1) 65 or older 121 12.0 (11.5) 6.0 (5.5) 9.0 (6.2) 27.0 (17.8) t test P = 0.038* P = 0.031* P = 0.486 P = 0.310 Sex Male 142 13.1 (11.3) 5.6 (5.1) 9.3 (6.0) 27.9 (17.0) Female 94 14.2 (12.0) 4.9 (4.6) 9.4 (6.2) 28.4 (18.2) t test P = 0.476 P = 0.255 P = 0.841 P = 0.824 Duration of diabetes \ 10 years 93 13.2 (11.9) 4.4 (4.4) 9.1 (5.5) 26.7 (17.7) 10 years or longer 142 13.8 (11.4) 5.8 (5.1) 9.4 (6.3) 29.0 (17.3) Unknown 1 7.0 (–) 15.0 (–) 25.0 (–) 47.0 (–) t test (unknown excluded) P = 0.698 P = 0.030* P = 0.732 P = 0.328 Frequency of dose: injection None 118 11.0 (9.4) 4.9 (4.8) 8.8 (6.0) 24.6 (15.6) 1/week 40 12.3 (11.0) 2.7 (3.5) 9.4 (5.4) 24.3 (14.9) 1/day 41 18.3 (13.9) 6.6 (4.3) 9.3 (6.1) 34.2 (18.3) 2/day 8 10.4 (9.4) 7.1 (5.0) 9.4 (5.7) 26.9 (13.9) Diabetes Ther (2018) 9:1001–1019 1013 Table 5 continued Patients’ backgrounds N DTBQ scores Implementation burden Flexibility burden BG control burden Total burden Category Mean (SD) Mean (SD) Mean (SD) Mean (SD) 3/day 8 19.5 (13.3) 8.4 (6.1) 14.1 (8.2) 42.0 (25.0) 4/day 21 19.4 (13.9) 8.4 (5.5) 10.5 (6.5) 38.3 (20.3) ANOVA (none excluded) P = 0.082 P \ 0.001*** P = 0.306 P = 0.011* Frequency of dose: OHA None 21 15.9 (13.4) 8.7 (5.2) 10.8 (5.9) 35.4 (19.3) 1/week 40 9.4 (10.2) 2.2 (2.8) 7.3 (4.4) 18.9 (14.4) 1/day 62 11.9 (11.3) 5.0 (4.8) 9.1 (6.8) 25.9 (16.8) 2/day 48 13.8 (11.0) 5.4 (4.6) 9.2 (6.2) 28.4 (17.7) 3/day or more 65 16.6 (11.6) 6.4 (5.1) 10.4 (5.9) 33.4 (16.5) ANOVA (none excluded) P = 0.009** P \ 0.001*** P = 0.093 P \ 0.001*** Diabetic neuropathy No 147 13.0 (11.1) 4.8 (4.7) 9.0 (5.8) 26.8 (16.4) Yes 52 14.6 (12.8) 5.9 (4.9) 9.4 (6.1) 30.0 (18.9) Unknown 37 14.0 (11.9) 6.4 (5.6) 10.4 (6.9) 30.7 (19.2) t test (unknown excluded) P = 0.379 P = 0.154 P = 0.681 P = 0.259 Diabetic nephropathy No 174 13.1 (11.4) 5.0 (4.9) 8.8 (6.1) 26.9 (17.2) Yes 62 14.7 (12.1) 6.2 (5.0) 10.7 (5.8) 31.6 (17.9) t test P = 0.355 P = 0.081 P = 0.039* P = 0.069 Diabetic retinopathy No 150 13.0 (11.9) 4.8 (4.7) 8.7 (5.8) 26.4 (17.3) Yes 61 14.5 (10.0) 6.7 (4.9) 10.7 (6.1) 31.9 (16.4) Unknown 25 14.2 (13.3) 5.2 (5.4) 9.8 (6.7) 29.2 (20.0) t test (unknown excluded) P = 0.378 P = 0.007** P = 0.031* P = 0.036* ANOVA analysis of variance, BG blood glucose, BID? twice a day or more, DTBQ Diabetic Treatment Burden Ques- tionnaire, OHA oral hypoglycemic agent, QD once daily, QW once weekly *P \ 0.05, **P \ 0.01, ***P \ 0.001 control, hypoglycemic event, frequency of injection or OHA, higher HbA1c (C 7.0%), with missed dose: injection or OHA, communication diabetic nephropathy or retinopathy, with BG with physician on diabetic treatment). DTBQ controlled poorly, with hypoglycemic events, BG control burden score was high in the fol- frequently missing dose of injection or OHA, lowing patients: frequently missing dose of less communication with their physicians. 1014 Diabetes Ther (2018) 9:1001–1019 treatment. In particular, the flexibility burden DISCUSSION score rose significantly as the frequency of dose increased. Implementation burden score also We developed DTBQ, a patient-administered increased with order of dosing frequency, and questionnaire with 18 questions, to measure the was higher for injection than OHA. treatment burden caused by pharmacotherapy HbA1c was significantly associated with total for T2DM. The reproducibility and validity of burden score. Total burden score was higher in DTBQ were evaluated through statistical analy- patients with higher HbA1c (C 7.0%). Of the sis of data collected using the questionnaire DTBQ subscales, BG control burden score and from patients with T2DM under pharma- flexibility burden score also showed significant cotherapy. This included 236 patients as the differences with HbA1c. BG control reported by validity analysis set and 47 patients as the patients was significantly associated with DTBQ reproducibility analysis set. scores as well. These results suggested that Weighted kappa coefficients were calculated patients with worse BG control felt a heavier as an index of reproducibility for the DTBQ item burden for treatment of diabetes. scores, with all 18 questions showing repro- Age was not significantly associated with ducibility high enough for practical use. ICCs total burden score. However, among the DTBQ were also calculated as an index of repro- subscales, implementation burden score was ducibility for the DTBQ subscale scores and significantly higher in younger patients total burden score. All coefficients were 0.8 or (\ 65 years). Younger patients might feel heav- higher, showing high reproducibility of three ier implementation burdens because of their subscales and total burden score. greater job demands. On the other hand, flexi- Factor analysis of the DTBQ item scores bility burden score was significantly higher in without rotation showed a unidimensional older patients (C 65 years). feature of DTBQ, which supported interpreting Hypoglycemic events were significantly the total score of all 18 questions as ‘‘total bur- associated with total burden score and DTBQ den’’. On the basis of the result of the three- subscale scores. All scores were significantly factor model with varimax rotation, a simple higher in patients with hypoglycemic events. structure was obtained, and three subscales were Out of the six types of diabetic treatment, the determined and designated as follows: ‘‘imple- percentage of hypoglycemic events and all mentation burden’’, ‘‘flexibility burden’’, and DTBQ subscale scores were highest for Injec- ‘‘blood control (BG) control burden’’. tion/BID?, which reflected the association There were significant associations between between hypoglycemia and DTBQ scores. This type of diabetic treatment and DTBQ total bur- suggests that patients with hypoglycemic events den score. In terms of administration route would probably feel a heavier treatment burden (injection vs. OHA), the score was higher for for controlling BG. injection than oral administration. For the fre- Daily frequency of dosing non-diabetic quency of dose (QW, QD, BID?), the score agents was not significantly associated with any increased with order of dosing frequency: QW, DTBQ score. This result suggested that DTBQ QD, and BID?. Comparing the means of total measured the treatment burden caused by burden score among the type of diabetic treat- pharmacotherapy for diabetes only, as ment (combination of administration route and intended. dose frequency), OHA/BID? (33.9) was almost Frequency of missed dose of diabetic medi- the same as Injection/QD (34.2), and Injection/ cation (injection or OHA) was significantly QW (24.3) was smaller than OHA/BID?. These associated with the total burden score and all results suggested that the treatment burden of DTBQ subscale scores. All scores were signifi- Injection/QW was lighter than that of OHA/ cantly higher in patients with a missed injec- BID?. tion or OHA dose. As the frequency of dosing The DTBQ subscale scores were also signifi- diabetic medication increased, the frequency of cantly associated with type of diabetic missed dose and all subscale scores also Diabetes Ther (2018) 9:1001–1019 1015 increased simultaneously, which reflected the clearly preferred to QD. The questions in DTBQ association between the frequency of missed directly address factors affecting patients’ pref- dose and the DTBQ scores. erence, so DTBQ can contribute to a better The treatment burden in patients with dia- understanding of patient preference. betic retinopathy was higher than that in This study has some limitations. In the patients without diabetic retinopathy. In the validity analysis set of this study, mean HbA1c patients with diabetic retinopathy, 41 out of 61 was 6.98, which indicated that the study sub- subjects had injection therapy compared with jects controlled BG better than the general 63 out of 150 subjects in the patients without T2DM population. Another study should be diabetic retinopathy. Duration of illness of conducted to evaluate the validity of DTBQ for patients with diabetic retinopathy is relatively poorly controlled T2DM patients. Because the longer and insulin therapy is necessary for type of diabetic treatment was not randomized, many patients. Since injection therapy has a results based on comparisons of DTBQ scores higher burden overall, there was a statistically with treatment type should not be interpreted significant difference between the groups. as interventional effects. The treatment burden DTSQ satisfaction score was negatively cor- should be measured from various points of related with all DTBQ scores. The expected view, and there are some aspects that DTBQ results suggested that DTBQ validly measured does not cover. For example, side effects other diabetic treatment burden. Moreover, the than hypoglycemia (e.g., weight gain, gas- results of pairwise comparisons of DTBQ/DTSQ trointestinal symptoms) and financial aspects scores among types of diabetic treatment (e.g., cost of drug and device) are not included. (Table S4 in the electronic supplementary Other questionnaires measuring the uncovered material) showed that the mean DTBQ total aspects should be used together depending on burden score and DTBQ satisfaction score were the purpose, such as OHA-Q [36]. The original inversely related to the types of diabetic treat- version of DTBQ was written in Japanese; we ment. This indicated that the heavier burden of translated the questionnaire into English for diabetic treatment might worsen treatment this publication. If there are plans to use the satisfaction. English version with patients, a validation study In the studies comparing QW and BID for- will be needed. mulations of GLP-1 receptor agonist, better It has been already shown that improving treatment satisfaction was observed in patients treatment satisfaction and quality of life of treated with QW medication [24, 25]. Looking patients has a positive impact on their adher- at another therapeutic area, some observa- ence and BG control [37, 38]. In this study, tional/database studies on pharmacotherapy of DTBQ total burden score correlated negatively osteoporosis concluded that QW dosing was with DTSQ satisfaction score. Combining these associated with better adherence and persis- results, it is expected that increase of the tence than QD dosing [26–28]. These results treatment burden leads to worsening BG con- were consistent with the differences among trol. In ‘‘Standards of Medical Care in Dia- type of diabetic treatment observed in the betes—2017’’ released by the American DTBQ scores. Diabetes Association [39], the following state- Recently, patient preference of diabetic ments were included: treatment has been studied and reported • Treatment decision should be timely, rely on intensively [29–35]. For OHAs, a daily dosing evidence-based guidelines, and be made col- laboratively with patients based on individ- schedule was an important factor in addition to efficacy, safety, and cost. For injections, dosing ual preferences, prognosis, and comorbidities. frequency was the factor which most affected patient preference among all characteristics of • Providers should consider the burden of GLP-1 receptor agonists, including type of treatment and self-efficacy of patients when delivery system, nausea, hypoglycemia, BG recommending treatment. change, and dosing frequency where QW was 1016 Diabetes Ther (2018) 9:1001–1019 These statements reflect the importance of Disclosures. Hitoshi Ishii has received patient preference and treatment burden for research grants from Eli Lilly Japan and hono- determining the most appropriate treatment raria from Ono Pharmaceutical, Shionogi, Mit- option for diabetes. DTBQ is very much in the subishi Tanabe Pharma, Arkray, MSD, Taisho same direction indicated by these statements, Toyama Pharmaceutical, Becton–Dickinson and it enables quantitative measurement of the Japan, Novo Nordisk Pharma, Daiichi Sankyo, treatment burden caused by pharmacotherapy. Takeda Pharmaceutical, Astellas Pharma, and Measuring the treatment burden on patients Boehringer Ingelheim Japan, and lecture and/or with poor adherence would help to better consultant fees from Takeda Pharmaceutical, Eli understand their reasons, and some counter- Lilly Japan, Sanofi, MSD, Astellas Pharma, measures might be taken to improve their Novartis Pharma, Mitsubishi Tanabe Pharma, adherence. Daiichi Sankyo, Ono Pharmaceutical, AstraZe- In conclusion, on the basis of the above neca, Taisho Toyama Pharmaceutical, Shionogi, results and the considerations discussed, DTBQ Kowa Pharmaceutical, Boehringer Ingelheim has adequate reproducibility and validity as a Japan, Novo Nordisk Pharma, Sumitomo measurement scale for treatment burden on Dainippon Pharma, Kyowa Hakko Kirin, Ter- T2DM patients. umo Co., Bristol-Myers Squibb Co., Sanwa Kagaku Kenkyusho Co. Ltd., Fujirebio Inc., Kowa Pharmaceutical, Pfizer, and Teijin Pharma. Hiroki Shin has no conflict of interests. ACKNOWLEDGEMENTS Takahiro Tosaki has received lecture fees from Eli Lilly Japan, Takeda Pharmaceutical, and MSD. Tatsuya Haga has received lecture fees Funding. This study was funded by Eli Lilly from Eli Lilly Japan and Novo Nordisk Pharma. Japan (Kobe, Japan). No funding was received Yoshiki Nakajima has no conflict of interests. for the article processing charges and medical Toshihiko Shiraiwa has received research grants writing/editorial assistance. from Novo Nordisk Pharma, Sanofi, Takeda Pharmaceutical, AstraZeneca, Boehringer Ingel- Authorship. All named authors meet the heim Japan, and Quintiles and lecture fees from International Committee of Medical Journal Takeda Pharmaceutical and Sanofi. Nobuaki Editors (ICMJE) criteria for authorship for this Watanabe has received research grants from article, take responsibility for the integrity of Novo Nordisk Pharma and Astellas Pharma. the work as a whole, and have given their Miyuki Koizumi has received lecture fees from approval for this version to be published. Takeda Pharmaceutical. Hiroki Nakajima has Hitoshi Ishii was responsible for study design, received lecture fees from Astellas Pharma and interpretation of the results, and drafting this Takeda Pharmaceutical. Sadanori Okada has manuscript. All authors contributed to the data received lecture fees from MSD, Eli Lilly Japan, collection, reviewing, and final approval of this Takeda Pharmaceutical, Ono Pharmaceutical, manuscript. All authors had full access to all of Novo Nordisk Pharma, AstraZeneca, Mitsubishi the data in this study and take complete Tanabe Pharma, Sumitomo Dainippon Pharma, responsibility for the integrity of the data and Boehringer Ingelheim Japan, Sanofi, and Ark- accuracy of the data analysis. ray. Tsuyoshi Mashitani has received lecture Medical Writing and Editorial Assis- fees from MSD, Eli Lilly Japan, Boehringer tance. The authors thank Eisei Oda and Junko Ingelheim Japan, Kissei Pharmaceutical, Takeda Yamada in Medical TOUKEI Corp. (Tokyo, Pharmaceutical, Ono Pharmaceutical, Novo Japan) for data management, statistical analysis, Nordisk Pharma, Daiichi Sankyo, Mitsubishi and medical writing. No funding was received Tanabe Pharma, Sanofi, and AstraZeneca. for this. Takako Mohri has received lecture fees from Novo Nordisk Pharma, Takeda Pharmaceutical, Boehringer Ingelheim Japan, and Eli Lilly Japan. Diabetes Ther (2018) 9:1001–1019 1017 US diabetes care, 1999–2010. N Engl J Med. Yasuhiro Akai has received lecture fees from 2013;368(17):1613–24. MSD, Astellas Pharma, Otsuka Pharmaceutical, Ono Pharmaceutical, Kinden, Kissei Pharma- 3. Cramer JA. A systematic review of adherence with ceutical, Sanofi, Daiichi Sankyo, Sumitomo medications for diabetes. Diabetes Care. 2004;27(5):1218–24. Dainippon Pharma, Mitsubishi Tanabe Pharma, Chugai Pharmaceutical, Teijin Pharma, Torii 4. Donnelly LA, Morris AD, Evans JM, DARTS/MEMO Pharmaceutical, Eli Lilly Japan, Boehringer collaboration. Adherence to insulin and its associ- Ingelheim Japan, Novo Nordisk Pharma, Baxter, ation with glycaemic control in patients with type 2 diabetes. QJM. 2007;100(6):345–50. and Bristol-Myers Squibb. 5. Farmer AJ, Rodgers LR, Lonergan M, et al. Adher- Compliance with Ethics Guidelines. A ence to oral glucose-lowering therapies and associ- common protocol was approved at institutional ations with 1-year HbA1c: a retrospective cohort review boards in Nara Medical University and analysis in a large primary care database. Diabetes Care. 2016;39(2):258–63. TDE Healthcare Corporation TOSAKI Clinic for Diabetes and Endocrinology. All procedures 6. Bailey CJ, Kodack M. Patient adherence to medica- performed in studies involving human partici- tion requirements for therapy of type 2 diabetes. Int pants were in accordance with the ethical J Clin Pract. 2011;65(3):314–22. standards of the institutional and/or national ´ ´ ´ 7. Garcıa-Perez LE, Alvarez M, Dilla T, Gil-GuillenV, research committee and with the 1964 Decla- Orozco-Beltran D. Adherence to therapies in ration of Helsinki and its later amendments or patients with type 2 diabetes. Diabetes Ther. comparable ethical standards. Informed con- 2013;4(2):175–94. sent was obtained from all individual partici- 8. Kirkman MS, Rowan-Martin MT, Levin R, et al. pants included in the study. Determinants of adherence to diabetes medica- tions: findings from a large pharmacy claims data- Data Availability. The datasets generated base. Diabetes Care. 2015;38(4):604–9. and/or analyzed during the current study are 9. Polonsky WH, Henry RR. Poor medication adher- available from the corresponding author on ence in type 2 diabetes: recognizing the scope of the reasonable request. problem and its key contributors. Patient Prefer Adherence. 2016;10:1299–307. Open Access. This article is distributed 10. Peron EP, Ogbonna KC, Donohoe KL. Antidiabetic under the terms of the Creative Commons medications and polypharmacy. Clin Geriatr Med. Attribution-NonCommercial 4.0 International 2015;31(1):17–27. License (http://creativecommons.org/licenses/ by-nc/4.0/), which permits any non- 11. Schernthaner G. Fixed-dose combination therapies in the management of hyperglycaemia in type 2 commercial use, distribution, and reproduction diabetes: an opportunity to improve adherence and in any medium, provided you give appropriate patient care. Diabet Med. 2010;27(7):739–43. credit to the original author(s) and the source, provide a link to the Creative Commons license, 12. Blonde L, San Juan ZT. Fixed-dose combinations for treatment of type 2 diabetes mellitus. Adv Ther. and indicate if changes were made. 2012;29(1):1–13. 13. Polonsky WH, Fisher L, Hessler D, Bruhn D, Best JH. Patient perspectives on once-weekly medications REFERENCES for diabetes. Diabetes Obes Metab. 2011;13(2):144–9. 1. Thrasher J. Pharmacologic management of type 2 14. Kaku K. First novel once-weekly DPP-4 inhibitor, diabetes mellitus: available therapies. Am J Cardiol. trelagliptin, for the treatment of type 2 diabetes 2017;120(1S):S4–16. mellitus. Expert Opin Pharmacother. 2015;16(16):2539–47. 2. Ali MK, Bullard KM, Saaddine JB, Cowie CC, Imperatore G, Gregg EW. Achievement of goals in 1018 Diabetes Ther (2018) 9:1001–1019 15. Evans PM, Bain SC. Omarigliptin for the treatment 27. Kishimoto H, Maehara M. Compliance and persis- of type 2 diabetes mellitus. Expert Opin Pharma- tence with daily, weekly, and monthly bisphos- cother. 2016;17(14):1947–52. phonates for osteoporosis in Japan: analysis of data from the CISA. Arch Osteoporos. 2015;10:231. 16. Brunton S, Davidson JA. Exenatide once weekly: a review of pharmacology and treatment considera- 28. Iglay K, Cao X, Mavros P, Joshi K, Yu S, Tunceli K. tions in type 2 diabetes. Clin Ther. Systematic literature review and meta-analysis of 2016;38(3):582–94. medication adherence with once-weekly versus once-daily therapy. Clin Ther. 2015;37(8):1813–21. 17. Burness CB, Scott LJ. Dulaglutide: a review in type 2 diabetes. BioDrugs. 2015;29(6):407–18. 29. Hauber AB, Han S, Yang JC, et al. Effect of pill burden on dosing preferences, willingness to pay, 18. Qiao Q, Ouwens MJ, Grandy S, Johnsson K, Kostev and likely adherence among patients with type 2 K. Adherence to GLP-1 receptor agonist therapy diabetes. Patient Prefer Adherence. 2013;7:937–49. administered by once-daily or once-weekly injec- tion in patients with type 2 diabetes in Germany. 30. Hauber AB, Tunceli K, Yang JC, et al. A survey of Diabetes Metab Syndr Obes. 2016;9:201–5. patient preferences for oral antihyperglycemic therapy in patients with type 2 diabetes mellitus. Diabetes Ther. 2015;6(1):75–84. 19. Nguyen H, Dufour R, Caldwell-Tarr A. Glucagon- like peptide-1 receptor agonist (GLP-1RA) therapy adherence for patients with type 2 diabetes in a 31. Gelhorn HL, Poon JL, Davies EW, Paczkowski R, medicare population. Adv Ther. 2017;34(3):658–73. Curtis SE, Boye KS. Evaluating preferences for pro- files of GLP-1 receptor agonists among injection- 20. Polonsky WH, Anderson BJ, Lohrer PA, et al. naı¨ve type 2 diabetes patients in the UK. Patient Assessment of diabetes-related distress. Diabetes Prefer Adherence. 2015;9:1611–22. Care. 1995;18(6):754–60. 32. Hauber AB, Nguyen H, Posner J, Kalsekar I, Ruggles 21. Polonsky WH, Fisher L, Earles J, et al. Assessing J. A discrete-choice experiment to quantify patient psychosocial distress in diabetes: development of preferences for frequency of glucagon-like peptide- the diabetes distress scale. Diabetes Care. 1 receptor agonist injections in the treatment of 2005;28(3):626–31. type 2 diabetes. Curr Med Res Opin. 2016;32(2):251–62. 22. Dennick K, Sturt J, Speight J. What is diabetes dis- tress and how can we measure it? A narrative review 33. Gelhorn HL, Bacci ED, Poon JL, Boye KS, Suzuki S, and conceptual model. J Diabetes Complicat. Babineaux SM. Evaluating preferences for profiles of 2017;31(5):898–911. glucagon-like peptide-1 receptor agonists among injection-naive type 2 diabetes patients in Japan. 23. Pouwer F, Snoek FJ, van der Ploeg HM, Heine RJ, Patient Prefer Adherence. 2016;10:1337–48. Brand AN. A comparison of the standard and the computerized versions of the Well-being Ques- 34. Qin L, Chen S, Flood E, et al. Glucagon-like peptide- tionnaire (WBQ) and the Diabetes Treatment Sat- 1 receptor agonist treatment attributes important to isfaction Questionnaire (DTSQ). Qual Life Res. injection-experienced patients with type 2 diabetes 1998;7:33–8. mellitus: a preference study in Germany and the United Kingdom. Diabetes Ther. 2017;8(2):335–53. 24. Best JH, Boye KS, Rubin RR, Cao D, Kim TH, Peyrot M. Improved treatment satisfaction and weight-re- 35. Qin L, Chen S, Flood E, et al. Glucagon-like peptide- lated quality of life with exenatide once weekly or 1 receptor agonist treatment attributes important to twice daily. Diabet Med. 2009;26(7):722–8. injection-naı¨ve patients with type 2 diabetes mel- litus: a multinational preference study. Diabetes 25. Reaney M, Yu M, Lakshmanan M, Pechtner V, van Ther. 2017;8(2):321–34. Brunt K. Treatment satisfaction in people with type 2 diabetes mellitus treated with once-weekly 36. Ishii H, Oda E. Reproducibility and validity of a dulaglutide: data from the AWARD-1 and AWARD- satisfaction questionnaire on hypoglycemic agents: 3 clinical trials. Diabetes Obes Metab. the Oral Hypoglycemic Agent Questionnaire (OHA- 2015;17(9):896–903. Q). Diabetol Int. 2012;3(3):152–63. 26. Confavreux CB, Canoui-Poitrine F, Schott AM, 37. Roborel de Climens A, Tunceli K, Arnould B, et al. Ambrosi V, Tainturier V, Chapurlat RD. Persistence Review of patient-reported outcome instruments measuring health-related quality of life and satis- at 1 year of oral antiosteoporotic drugs: a prospec- faction in patients with type 2 diabetes treated with tive study in a comprehensive health insurance database. Eur J Endocrinol. 2012;166(4):735–41. oral therapy. Curr Med Res Opin. 2015;31(4):643–65. Diabetes Ther (2018) 9:1001–1019 1019 38. Ishii H, Anderson JH Jr, Yamamura A, Takeuchi M, 39. American Diabetes Association. Promoting health Ikeda I. Improvement of glycemic control and and reducing disparities in populations. Diabetes quality-of-life by insulin lispro therapy: assessing Care. 2017;40(Suppl 1):S6–10. benefits by ITR-QOL questionnaires. Diabetes Res Clin Pract. 2008;81(2):169–78.

Journal

Diabetes TherapySpringer Journals

Published: Mar 29, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off