FormalPara Key Summary Points

Studies that assess the patient-reported outcomes (PROs) of persons with type 2 diabetes mellitus (T2DM) initiating injectable glucose-lowering medications in routine clinical practice are limited.

TROPHIES is a 24-month prospective observational study performed in France, Germany, and Italy in patients with T2DM who initiated their first injectable glucose-lowering medication with once-weekly dulaglutide or once-daily liraglutide.

The perspectives of patients based on a diversified panel of generic and disease-specific PRO measures at the time of study enrollment are described.

At baseline, patients initiating dulaglutide or liraglutide rated their quality of life and other health outcomes; the results corresponded to a mean EQ-5D-5L index of 0.84 and 0.83, a mean EQ-VAS score of 67.5 and 67.5, a mean IW-SP score of 59.8 and 61.3, a mean DTSQs score of 24.6 and 25.8, a mean DPM Life Productivity score of 78.6 and 79.5, and a mean DPM Work Productivity score of 87.5 and 86.8, respectively.  

In clinical practice, patients with T2DM who needed treatment intensification to improve glycemic control and were initiating their first injectable therapy with either dulaglutide or liraglutide generally reported similar quality of life ratings.

The information from this varied panel of PRO instruments in a real-world setting complements clinical outcomes data.

Digital Features

This article is published with digital features to facilitate understanding of the article. You can access the digital features on the article’s associated Figshare page. To view digital features for this article go to https://doi.org/10.6084/m9.figshare.12783287.

Introduction

The management of type 2 diabetes mellitus (T2DM) through diet, lifestyle, and medication has consequences for the patient’s quality of life. With the growing number and types of glucose-lowering medications (GLMs), choosing an appropriate T2DM medication can complicate management and represent a burden for patients [1, 2]. T2DM is a progressive disease and maintaining glycemic control over time often requires treatment intensification by switching to injectable therapy. Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are typically the first injectable therapy recommended for T2DM, as they offer improved glycemic control and other health benefits [1,2,3]. Dulaglutide (Trulicity®, Eli Lilly and Company, USA; approved in the European Union in 2014) and liraglutide (Victoza®, Novo Nordisk, Denmark; approved in the European Union in 2010) are injectable GLP-1 RAs approved for T2DM treatment, and both are associated with positive clinical outcomes [4,5,6,7,8,9].

Collected real-world evidence can complement randomized clinical trials, providing, for example, longer-term data on treatment trends, patient adherence, durability of clinical outcomes, and economic data, all of which are fundamental for healthcare decision makers. Real-world studies involving patients commencing their first injectable therapy with dulaglutide or liraglutide can help us to understand the impact of prescribing GLP-1 RAs on T2DM clinical outcomes in routine practice. Patient-reported outcome (PRO) measures can complement clinical outcomes and, beyond efficacy and safety data, enrich our understanding of the patients’ perspective of GLP-1 RA use in routine clinical practice [10, 11].

Although PROs for patients taking dulaglutide and liraglutide have previously been reported in clinical trials, real-world evidence is limited [11,12,13,14,15,16,17]. We studied the use of these GLMs in a single prospective observational study, thereby addressing the need for additional real-world evidence. One of the objectives of the study (TROPHIES; The Real-world Observational Prospective study of Health outcomes with dulaglutIde and liraglutide in type 2 diabeteS patients) is to assess PROs regarding overall quality of life, treatment satisfaction, and productivity in patients initiating their first GLP-1 RA injectable—either dulaglutide or liraglutide—as part of T2DM therapy. At the commencement of the study, dulaglutide and liraglutide were the most prominently used GLP-1 RAs in many European countries. Clinical information at baseline has been presented previously [18, 19]. This paper focuses on the responses to the PRO questionnaires collected at the time of patient enrollment, providing descriptive information on patient populations initiating dulaglutide or liraglutide as the first GLP-1 RA treatment.

Methods

Patients and Study Design

TROPHIES is a 24-month prospective observational study conducted in two cohorts of T2DM patients initiating injectable GLP-1 RA treatment with dulaglutide or liraglutide. Male or female patients aged ≥ 18 years were eligible to participate if they were naive to injectable treatment and were scheduled to begin their first injectable antihyperglycemic treatment with either dulaglutide once weekly or liraglutide once daily according to the approved label and at the discretion of the diabetes health care provider in the respective country. The patients were enrolled from three European countries: France, Germany, and Italy. Data were collected at baseline and will continue to be collected at routine clinical care visits approximately 6, 12, 18, and 24 months after initiating GLP-1 RA treatment (Fig. S1 in the Electronic supplementary material, ESM).

Study Outcomes

The primary objective of the TROPHIES study is to estimate the time spent on the first GLP-1 RA (initiated at baseline) without a significant treatment change (time to first significant treatment change) due to treatment- or diabetes-related factors. One of the secondary objectives of the TROPHIES study is to describe the PROs associated with the treatment of diabetes with dulaglutide or liraglutide, including health-related quality of life (HR-QoL), treatment satisfaction, and productivity.

Study Procedures and Data Management

Dulaglutide and liraglutide are subcutaneously administered as directed by the respective Summary of Product Characteristics [20, 21]. The recommended dose for dulaglutide is 0.75 mg once weekly as monotherapy and 1.5 mg once weekly when used as add-on therapy. The 0.75 mg once-weekly dose can be considered a starting dose in add-on therapy for potentially vulnerable populations. The liraglutide starting dose is 0.6 mg per day. After at least one week, the liraglutide dose is increased to 1.2 mg per day. This can be further increased to 1.8 mg for patients that are expected to benefit from a dose increase, and based on the clinical response.

At the first study visit, information was collected from participating patients and from their medical records over the previous 6-month period. This information included demographic data, clinical history (including diabetes duration, previous glycated hemoglobin (HbA1c) level, concomitant diseases, previous treatment, and non-antihyperglycemic medications). Data on resource use for diabetes during the last 6 months were also collected. Electronic case report forms were used for data entry. The HR-QoL and other PRO instruments described in the following sections were administered and completed via paper questionnaires during the first study visit, prior to initiating treatment with dulaglutide or liraglutide.

PRO Measures

Five PRO measures were administered at baseline and at each postbaseline visit (Table 1). These included the Diabetes Treatment Satisfaction Questionnaire Status (DTSQs) [22, 23], EQ-5D-5L and the EQ-5D Visual Analog Scale (EQ-VAS) [24, 25], the Impact of Weight on Self-Perceptions Questionnaire (IW-SP) [26], and the Diabetes Productivity Measure (DPM) [27].

Table 1 Overview of the PRO measures administered at baseline in the TROPHIES study
Full size table

DTSQs is used to evaluate satisfaction with diabetes treatment regimens, as well as concerns about hyperglycemia and hypoglycemia [22, 23]. Six items are measured (satisfied with current treatment; feel convenient about recent treatment; feel flexible about recent treatment; satisfied with diabetes understanding; recommend present treatment to others; continue with present treatment) and summed to provide a treatment satisfaction score on a scale of 0–36. Higher scores indicate greater satisfaction. Two items concerning times of unacceptably high and low blood glucose are rated from 0 (none of the time) to 6 (most of the time). These items are scored separately from the six treatment satisfaction items and from each other. Here, lower scores indicate blood glucose levels closer to the ideal and higher scores indicate problems with blood glucose levels.

The EQ-5D-5L is a descriptive instrument that is used to measure overall health status [25]. A weighted summary index score is derived for five dimensions (mobility; self-care; usual activities; pain/discomfort; and anxiety/depression), with each dimension having five response levels (1 = no problems; 2 = slight problems; 3 = moderate problems; 4 = severe problems; 5 = unable to/extreme problems). The index score is reported on a scale 0 to 1, whereby a score of 1 indicates perfect health, 0 indicates death, and negative scores represent states that are worse than death. The EQ-VAS is completed in conjunction with the EQ-5D-5L and allows patients to self-rate their health status on a scale of 0 to 100, where 0 is the worst imaginable health and 100 is the best imaginable health state.

The IW-SP questionnaire is designed to assess the patients’ self-perception relating to their body weight [26]. Three items are measured (feel unhappy with appearance due to weight; feel self-conscious in public due to weight; feel unhappy due to comparing weight with others) on a 5-point scale ranging from 1 (always) to 5 (never). The raw scores are averaged and linearly transformed to produce a measure from 0 to 100. Higher raw scores and higher transformed scores correspond to better self-perception in relation to one's weight.

The DPM assesses life and work productivity impairments associated with diabetes [27]. The Life Productivity domain contains nine items (prevented from accomplishing important things; prevented from concentrating; interferes with ability to accomplish daily activities; takes longer than necessary to complete tasks; have trouble getting up and being active in the morning; have to limit your daily activities; accomplish less than would like to; too tired to accomplish as much as would like to; low blood sugar symptoms interfere with ability to perform daily activities). This domain is followed by the Work Productivity domain, which includes five items (difficulty performing work duties; difficulty controlling emotions with coworkers; feel less productive at work; miss work; need to reschedule meetings). All fourteen items are measured on a 5-point scale from 100 (never) to 0 (always). The scores are transformed on a 0- to 100-point scale, with higher scores indicating greater productivity. Two standalone “nonvalidated” items measuring short-term and long-term goals are included in the DPM and are reported separately. These items are on a 5-point scale from 100 (not at all) to 0 (extremely). Similarly, higher scores correspond to higher perceived life productivity.

Sample Size and Statistical Measures

The sample size of 350 patients per cohort in each country was deemed to be sufficient to provide a satisfactory estimate of the time until a significant treatment change occurred. This sample size was considered large enough to permit separate evaluations of patients initiating dulaglutide at the recommended 1.5 mg dose, to allow descriptive analyses of patients on the lower 0.75 mg dose, and to similarly assess the two liraglutide doses (1.2 mg and 1.8 mg).

Patients were described by treatment cohort (dulaglutide or liraglutide). Analysis was performed for the overall population (all countries pooled) and by each country. Absolute and relative frequencies were used to describe categorical variables. Descriptive summary statistics were presented for continuous variables, including number of patients, number of patients with missing data, mean ± standard deviation (SD), median quartiles, minimum values, and maximum values. PROs were scored according to the developer’s guidelines. Due to questionnaires not being completed, the sample sizes for each PRO varied and were less than the clinical data. The study protocol was approved by local institutional review boards, and all patients provided written informed consent. The study is executed in accordance with the Declaration of Helsinki 1964 and its later amendments and Guideline on Good Clinical Practices as well as country-specific requirements [28]. Eli Lilly & Company received approval to carry out the study at each study site before study initiation.

Results

Patients and Disease Characteristics at Baseline

In the TROPHIES study, 2181 patients were prescribed dulaglutide (N = 1130, mean (SD) 59 (11) years, 55% male) or liraglutide (N = 1051, 59 (11) years, 57% male) and were eligible for baseline clinical analysis (Table 2). The baseline demographics, clinical characteristics, and concomitant medication use were similar across patient cohorts. In both cohorts, the mean HbA1c levels at baseline exceeded the glycemic control targets reported by physicians in the study as their treatment objective. The mean difference between the baseline and target HbA1c values was statistically significantly different between the two cohorts (p = 0.017), with patients starting liraglutide further from the HbA1c goal at the start of treatment. In both cohorts, most of the patients had a difference between the baseline and target HbA1c levels of ≥ 1%. Around 86% of the patients were already receiving at least one oral GLM at the time of enrollment.

Table 2 Baseline demographic and clinical characteristics of patients initiating dulaglutide or liraglutide
Full size table

Diabetes-related conditions varied across patient cohorts (Table 2) [18]. The number of patients who had a macrovascular condition in the liraglutide cohort (25%) was significantly higher (p < 0.001) than the corresponding number in the dulaglutide (14%) cohort. There was no significant difference between the dulaglutide (20%) and liraglutide (19%) cohorts in the number of patients with at least one microvascular condition. The majority of patients had hyperlipidemia and hypertension.

PRO Measures

Treatment satisfaction was analyzed in a total of 2088 patients (Table 3). At baseline, patient satisfaction with the current treatment (before initiating dulaglutide or liraglutide) was similar for both cohorts. The overall mean (SD) DTSQs score was 25.17 (7.79), with patients initiating dulaglutide having lower satisfaction with current treatment (p < 0.001). In the TROPHIES study, data on the patients who perceived hyperglycemic and/or hypoglycemic episodes were also collected using the DTSQs questionnaire. Around 14% of patients reported no unacceptably high blood sugars, and around 50% of patients reported no unacceptably low blood sugars (Fig. 1). The perceived frequency of unacceptably high blood sugars [mean (SD) 3.30 (1.90)] was greater than the perceived frequency of unacceptably low blood sugars [mean (SD) 1.36 (1.69)] (Table 3). Around 70% of patients perceived experiencing three or more hyperglycemic episodes (Fig. 1).

Table 3 Mean patient-reported outcomes at baseline
Full size table
Fig. 1
figure 1

Results of DTSQs items concerning times of unacceptably high and low blood glucose at baseline. The proportions of patients reporting unacceptably low and/or high blood sugars rated from 0 (none of the time) to 6 (most of the time) are shown. The number of patients who, at this visit, were in the PRO population and answered at least one question on this questionnaire is given

Full size image

The EQ-5D-5L and the EQ-VAS questionnaires were completed by 2121 and 2119 respondents, respectively (Table 3). The weighted German index score [29] derived from the EQ-5D-5L questionnaire was high for both patient cohorts (a mean of 0.84 for dulaglutide and 0.83 for liraglutide). In the TROPHIES study, more than 50% of the patient population reported low scores (no problems) for four EQ-5D-5L dimensions: “mobility,” “self-care,” “usual activities,” and “anxiety/depression” (Fig. 2). “Pain/discomfort” was the most frequent patient-reported dimension. More than 60% of patients experienced slight to severe pain or discomfort. An overall mean score of 68 out of 100, which is the best imaginable health state, was measured with the EQ-VAS (Table 3).

Fig. 2
figure 2

Baseline results of the EQ-5D-5L quality of life questionnaire by dimension. The proportions of patients reporting quality of life statuses ranging from “no problem” to “unable to/extreme problems” are shown. The number of patients who were in the PRO population and answered at least one question on this questionnaire is also given (n)

Full size image

A total of 2108 respondents completed the IW-SP questionnaire (Table 3). The transformed scores were similar in the two cohorts (a mean of 60 for dulaglutide and 61 for liraglutide). In this assessment, one-third (approximately 35%) of respondents reported that they always or frequently feel unhappy with their appearance due to their weight (Fig. 3). Half of respondents reported rarely or never feeling self-conscious in public due to their weight (approximately 57%), and rarely or never feeling unhappy because they compared their weight with those of other people” (approximately 54%).

Fig. 3
figure 3

Baseline results of the IW-SP questionnaire by item. The number of patients who, at this visit, were in the PRO population and answered at least one question on this questionnaire is given (n)

Full size image

The number of respondents who completed the DPM Life and Work Productivity sections was 2114 and 1292, respectively (Table 3). The DPM scores were similar for the two cohorts. The work productivity scores were higher (87%) than the life productivity scores (79%). The mean (SD) score for the extent to which respondents felt that their diabetes prevented them from reaching their short-term and long-term goals was 84 (23) and 82 (24), respectively. More than 50% of patients reported no long-term or short-term goal impairments associated with T2DM (Fig. 4).

Fig. 4
figure 4

The DPM short-term and long-term goal results at baseline, rated from 100 (not at all) to 0 (extremely). The number of patients who, at this visit, were in the PRO population and answered this question is given (n)

Full size image

Discussion

TROPHIES is a real-world observational study assessing the responses of T2DM patients initiating an injectable GLP-1 RA for the first time. The data described herein provide an overview of the clinical characteristics and information collected from the five PRO measures for patients who were enrolled in the study across France, Germany, and Italy. Other information regarding the TROPHIES study has been published previously [18, 19].

Overall, the patient demographics and characteristics were similar for both patient cohorts, except for the difference between the baseline and target HbA1c values and the number of patients with at least one macrovascular condition. Most patients had a high BMI, reflective of a largely obese study population, and inadequately controlled diabetes (the overall mean HbA1c was 8.2%). These characteristics are similar to those reported in a previous study involving T2DM patients starting liraglutide treatment in a real-world clinical setting [11]. In the TROPHIES population, which had a mean diabetes duration of 8.5 years, HbA1c levels substantially exceeded the glycemic control targets set by physicians as their treatment objective, thereby supporting treatment intensification with GLP-1 RAs.

Treatment satisfaction is an important measure, as greater satisfaction is associated with better treatment compliance and improved persistence [30,31,32]. In this study, the DTSQs score of 25 out of 36 at baseline generally indicated moderate patient satisfaction with their current T2DM treatment prior to commencing the GLP-1 RA therapy. Patients initiating dulaglutide reported having less satisfaction with their current T2DM treatment than those initiating liraglutide. While the reasons for this difference are not known, the score for patients initiating dulaglutide was generally similar to that for patients initiating liraglutide. The DTSQs score indicated that there is potential to improve T2DM treatment satisfaction. This TROPHIES DTSQs baseline score is similar to what has previously been reported in other studies of T2DM patients before they started GLP-1 RA therapy [mean (SD) DTSQs score of 22.17 (7.64)] [11]. Of particular interest in this study is the mean score of 3.3 (out of 6.0) for the item regarding the time spent with unacceptably high blood glucose. The score indicated that patients perceived their hyperglycemic episode frequency was too high, consistent with a patient population in need of T2DM treatment intensification.

The EQ-5D-5L and self-reported health status (VAS) at baseline provide insight into the reality of living with diabetes and diabetes treatment, as perceived by the patient. Depending on the inclusion criteria and sample size for the study, EQ-5D-5L scores reported in several studies of patients with T2DM tend to be in the range of 0.7–0.8 [13, 33, 34]. The mean EQ-5D-5L index score found in this study is in line with such findings and indicates a moderate-to-positive health status [13, 35]. The EQ-VAS score, which indicates a moderate patient-rated health status, is also consistent with previously reported scores for patients with T2DM [34,35,36]. An assessment of the five individual EQ-5D-5L domains highlights the main challenges reported by patients: in this study, the majority of patients experienced problems with pain or discomfort, which explains why patients rated their health status as moderate according to the VAS scale.

Obesity is common among patients with T2DM. Obese patients with T2DM have been shown to more likely value weight loss than nonobese patients with T2DM [37]. Although published IW-SP data obtained in a real-world setting are limited, it is an important outcome in diabetes treatment [38]. In this TROPHIES study, moderate IW-SP scores were reported for both cohorts, indicating that patients were impacted by weight-related self-perceptions. The IW-SP indicated that the majority of patients were—at least sometimes—unhappy with their body weight, which is in line with previous descriptions of patients with T2DM in clinical trials [13]. This outcome is reasonable considering the high BMI [mean (SD) 33.9 (6.6) kg/m2] reported for this study population at baseline.

Published DPM data is limited; therefore, the additional information regarding patient work and life productivity obtained in this study will provide valuable insights into how diabetes impacts patients [39,40,41]. Overall, the Work Productivity and Life Productivity scores reported for the TROPHIES participants were 87% and 79%, respectively, and were similar to scores observed for patients facing T2DM treatment changes in a clinical trial setting [39]. Both this study and the previous clinical study showed that the work productivity scores were higher than the life productivity scores, perhaps indicating that T2DM impacts life productivity more than formal work productivity. The results from the standalone items on short-term and long-term goals showed that approximately 50% of study patients felt that diabetes did not impact their goals. The other 50% felt that the disease did impact their goals to some extent (ranging from slightly to extremely).

Limitations

This study may suffer from multiple limitations. These include the trial design, as patients are from a nonrandomized real-life setting, with potential confounding factors. Moreover, the overall patient population of the study may not be fully representative of T2DM cohorts due to potential physician-related and health care system-related selection processes, which could differ between countries. Not all patients completed all of the questionnaires. Among all of the PRO measures considered in this study, the questions relating to the DPM Work Productivity measure were most likely to be overlooked by patients. It is unclear whether this was because respondents believed that this part of the DPM only applied to working participants or because participants were not willing to complete the second page of the questionnaire.

Conclusion

The information gained from PROs in observational studies is important for understanding treatment benefits within a real-world setting. Such studies also provide the opportunity to evaluate patients’ perspectives in routine practice, which differs from a clinical trial setting. This diversified panel of both generic and disease-specific PRO instruments evaluated the perspectives of these real-world observational study T2DM patients who were starting GLP-1 RA therapy at the time of study enrollment. The data indicate that in clinical practice, patients with T2DM who need treatment intensification to improve glycemic control and who are initiating their first injectable therapy with either dulaglutide or liraglutide, generally reported similar health statuses and PROs at baseline. In this study, changes in the patient-reported outcomes over time will help us to understand patients’ perspectives regarding their quality of life and other parameters after starting the new GLP-1 RA treatment regimen in this real-world setting.