OBJECTIVE

While sodium–glucose cotransporter inhibitor (SGLTi) therapy has been evaluated in type 1 diabetes (T1D) trials, patient reactions to benefits and risks are unknown. Using established methodology, we evaluated patient preferences for different adjunct-to-insulin therapy options in T1D.

RESEARCH DESIGN AND METHODS

An online survey, completed by 701 respondents with T1D (231 U.S., 242 Canada, and 228 Germany), used conjoint analysis to present six hypothetical, masked, pairwise drug profile choices composed of different benefit-risk attributes and effect ranges. Data used in analyses were derived from actual phase 3 trials of a low-dose SGLTi (comparable to oral empagliflozin 2.5 mg q.d.), a high-dose SGLTi (comparable to oral sotagliflozin 400 mg q.d.), and an available adjunct-to-insulin therapy (comparable to subcutaneous pramlintide 60 μg t.i.d.).

RESULTS

Conjoint analysis identified diabetic ketoacidosis risk as most important to patients (23% relative score; z test, P < 0.05); ranked second were HbA1c reduction (14%), risk of severe hypoglycemia (13%), oral versus injectable treatment (12%), and risk of genital infection (12%). Next was risk of nausea (11%), followed by weight reduction (8%) and the risk of diarrhea (7%). A low-dose SGLTi drug profile was identified by conjoint analysis as the top patient preference (83% of participants; z test, P < 0.05) versus high-dose SGLTi (8%) or pramlintide (9%). Separate from conjoint analysis, when respondents were asked to choose their preferred adjunct-to-insulin therapy (masked to drug name/dose), 69%, 17%, 6%, and 9% of respondents chose low-dose SGLTi, high-dose SGLTi, pramlintide, and insulin therapy alone, respectively.

CONCLUSIONS

Low-dose SGLTi profile was the favored adjunct-to-insulin therapy by persons with T1D.

Recent diabetes management guidelines for the treatment of type 2 diabetes (T2D) from the American Diabetes Association and the European Association for the Study of Diabetes call for a patient-centric approach to enhance patient engagement in treatment decisions (1). Similar approaches are also important for persons with type 1 diabetes (T1D). It is therefore crucial to objectively evaluate patient preferences on attributes of new antidiabetes agents using well-established methodologies.

Patients living with T1D have relied, for nearly a century, solely on exogenous insulin administration as the mainstay of therapy for survival with no other reliable treatment options available to date. While advances in insulin delivery, continuous glucose monitoring (CGM), and improvement in insulin formulations have facilitated achieving glucose control, management remains complex and most patients are inadequately controlled (2,3). Adjunctive therapies to insulin in T1D are critically needed, but, as of 2018, only pramlintide has been approved in the U.S. Pramlintide’s efficacy profile in T1D, evidenced by a placebo-corrected HbA1c reduction in the range of 0.25–0.33% (2.7–3.6 mmol/mol), is offset by a twofold increase in the risk of severe hypoglycemia and the need for multiple daily injections, thereby limiting its clinical use (4,5).

In recent years, several molecules in the sodium–glucose cotransporter inhibitor (SGLTi) class including empagliflozin, dapagliflozin, and sotagliflozin have been studied in T1D in large phase 3 clinical trials (612). The positive benefit-risk profile of some of these agents, such as empagliflozin and dapagliflozin, has already been established in T2D, where these agents are indicated to improve glycemic control. Empagliflozin, specifically, has also been shown to reduce the risk of cardiovascular death in patients with T2D and established cardiovascular disease, and both empagliflozin and dapagliflozin appear to reduce hospitalization for heart failure and improve renal outcomes in T2D (1315).

Overall, SGLTi studies in T1D have shown a dose-dependent improvement of efficacy outcomes including benefits on HbA1c (−0.25% to −0.5% [−2.7 to −5.5 mmol/mol]), body weight (−2 to −3 kg), systolic blood pressure (−2 to −4 mmHg), and CGM-based glucose time in range (adding 1–3 h/day) (612). The efficacy observed with the SGLTi class in T1D has, however, come at an expense of an increased risk of genital infection (approximately a threefold increase vs. placebo) and, most importantly, an increased risk of diabetic ketoacidosis (DKA) of approximately three- to eightfold versus placebo for higher studied dose strengths (68). Interestingly, based on the empagliflozin T1D clinical development program, the lower empagliflozin 2.5 mg dose has not shown an increased risk for DKA versus placebo (8).

Indeed, concerns have been raised regarding the observed increased risk of DKA with the use of SGLTi in T1D despite risk mitigation strategies used in clinical trials (16,17). In this regard, post hoc analyses are being carried out to identify optimal T1D patient profiles for safe use of these agents. For example, the recent dapagliflozin and sotagliflozin approval by European regulators as adjunct to insulin therapy in T1D has been restricted in those with a BMI of ≥27 kg/m2. Some clinical experts have also advocated for the potential use of the lowest available dose strength of an SGLTi as another DKA risk mitigation strategy in T1D (18). However, little is known from objective analyses about the patient perspectives and preferences related to the overall SGLTi benefit-risk value proposition in T1D, including key efficacy and safety outcomes.

Given that patient voice is of central importance in therapeutic decision-making, we performed a survey-based study using a “discrete choice experiment” form of conjoint analysis, a validated and well-established methodology (19). Conjoint analysis was used to objectively evaluate the preferences of patients with T1D for adjunct-to-insulin therapy options including a low-dose SGLTi (comparable to oral empagliflozin 2.5 mg q.d.), a high-dose SGLTi (comparable to oral sotagliflozin 400 mg q.d.), and currently available adjunct-to-insulin therapy (subcutaneous pramlintide 60 μg t.i.d., approved for use in T1D in the U.S.).

Respondent Recruitment and Inclusion Criteria

Inclusion criteria for this survey required respondents to be adults diagnosed with T1D and residing in the U.S., Canada, or Germany. Respondents were recruited from dQ&A’s (San Francisco, CA) panel of patients with T1D engaged in a Web-based community. In Canada and Germany, additional respondents with HbA1c >8.0% (>64 mmol/mol) were screened and recruited from a general Web-based consumer panel to reflect the average standard of care currently achieved in terms of glycemic control in routine clinical care (e.g., similar to the T1D Exchange Registry) (2,3). Based on the nature of this survey, including the use of anonymous responses, there was no requirement for the involvement of an ethics board. Respondents received a nominal payment (U.S., US$10; Canada, Can$15; Germany, €12) for their dedicated time to complete the survey, which took a median of 15 min to complete.

Conjoint Analysis and Selection of Attributes/Levels

A “discrete choice experiment” form of conjoint analysis was used in this study to determine the preferences that patients with T1D implicitly place on specific treatment efficacy and safety attributes of adjunct-to-insulin therapy options. The list of selected attributes included treatment delivery (oral tablet vs. injectable), HbA1c reduction, weight reduction, risk of DKA, risk of severe hypoglycemia, risk of diarrhea, risk of nausea, and risk of genital infection. Various levels for each attribute were also defined, including options such as medication treatment form, various treatment effect sizes, or different safety risks for each attribute as applicable. The attributes and levels presented to the patient participants are summarized in Table 1.

Table 1

Attributes and levels used in conjoint analysis

AttributesLevels
Treatment form A pill once a day 
 Injections three times a day 
HbA1c reduction 0.23% (2.5 mmol/mol) 
 0.28% (3.1 mmol/mol) 
 0.46% (5.0 mmol/mol) 
Weight reduction 4.0 lbs (1.8 kg) 
 4.4 lbs (2 kg) 
 6.5 lbs (3 kg) 
Risk of DKA No change 
 5 times increase 
Risk of severe hypoglycemia No change 
 2 times increase 
Risk of diarrhea No change 
 2 times increase 
Risk of nausea No change 
 4 times increase 
Risk of genital infection No change 
 2 times increase 
 3 times increase 
AttributesLevels
Treatment form A pill once a day 
 Injections three times a day 
HbA1c reduction 0.23% (2.5 mmol/mol) 
 0.28% (3.1 mmol/mol) 
 0.46% (5.0 mmol/mol) 
Weight reduction 4.0 lbs (1.8 kg) 
 4.4 lbs (2 kg) 
 6.5 lbs (3 kg) 
Risk of DKA No change 
 5 times increase 
Risk of severe hypoglycemia No change 
 2 times increase 
Risk of diarrhea No change 
 2 times increase 
Risk of nausea No change 
 4 times increase 
Risk of genital infection No change 
 2 times increase 
 3 times increase 

Product attributes and levels were determined according to the published characteristics of drug profiles for oral empagliflozin 2.5 mg q.d., oral sotagliflozin 400 mg q.d., and the subcutaneous injection of pramlintide 60 μg t.i.d. The profile for the various drug options and levels of various attributes were adapted based on the safety and efficacy demonstrated in the placebo-controlled phase 3 clinical trial data for empagliflozin, sotagliflozin, and pramlintide evaluated as adjunct-to-insulin in T1D (4,7,8).

The research qualified the relative importance of the benefits and risks overall and among predefined subgroups of respondents. Expected respondent preference outcomes (also referred to elsewhere in the literature as “preference shares”), defined as the composite of attributes based on the medication efficacy and safety profiles from empagliflozin 2.5 mg q.d., sotagliflozin 400 mg q.d., and pramlintide 60 μg t.i.d., were also calculated.

Explicit Choice Assessments

The conjoint analysis was also complemented by an explicit choice questionnaire composed of the following: 1) A separate question in which respondents were asked to explicitly choose, masked to drug name and dose, from a drug profile comparable to a low-dose oral SGLTi referred to as “Medication X,” a high-dose oral SGLTi referred to as “Medication Y,” or an available injectable therapy referred to as “Medication Z” as an adjunct therapy in T1D (Table 2). In addition, a “none of the above” response option was also available to respondents should they have opted not to choose any of the three proposed medication profiles (X, Y, or Z). At the time of the design and conduct of this scientific survey, no SGLTi had gained regulatory approval for use in T1D. The survey was fielded from 3 October to 29 October 2018. 2) The likelihood of taking each of the three predefined choices (Medication X, Medication Y, and Medication Z as an adjunct to ongoing insulin therapy) was also characterized. The likelihood-to-adopt assessment was quantified based on a 4-point scale (definitely not/unlikely/likely/definitely) in addition to the estimation of a weighted likelihood score defined as 50% weight for the “definitely” responses plus a 25% weight for the “likely” responses.

Table 2

Masked drug profiles used in explicit choice analyses

Medication X (comparable to empagliflozin 2.5 mg q.d.)Medication Y (comparable to sotagliflozin 400 mg q.d.)Medication Z (comparable to pramlintide 60 μg t.i.d.)
Treatment form A pill once a day A pill once a day Injections three times a day 
HbA1c reduction 0.28% (3.1 mmol/mol) 0.46% (5.0 mmol/mol) 0.23% (2.5 mmol/mol) 
Weight reduction 4 lbs (1.8 kg) 6.5 lbs (3 kg) 4.4 lbs (2 kg) 
Risk of DKA No change 5 times increase No change 
Risk of severe hypoglycemia No change No change 2 times increase 
Risk of diarrhea No change 2 times increase No change 
Risk of nausea No change No change 4 times increase 
Risk of genital infection 2 times increase 3 times increase No change 
Medication X (comparable to empagliflozin 2.5 mg q.d.)Medication Y (comparable to sotagliflozin 400 mg q.d.)Medication Z (comparable to pramlintide 60 μg t.i.d.)
Treatment form A pill once a day A pill once a day Injections three times a day 
HbA1c reduction 0.28% (3.1 mmol/mol) 0.46% (5.0 mmol/mol) 0.23% (2.5 mmol/mol) 
Weight reduction 4 lbs (1.8 kg) 6.5 lbs (3 kg) 4.4 lbs (2 kg) 
Risk of DKA No change 5 times increase No change 
Risk of severe hypoglycemia No change No change 2 times increase 
Risk of diarrhea No change 2 times increase No change 
Risk of nausea No change No change 4 times increase 
Risk of genital infection 2 times increase 3 times increase No change 

Data Handling

In accordance with best practice, all incomplete responses, defined as surveys started by respondents but not fully completed and surveys completed in an unrealistically short time (defined as <7 min based on the minimum amount of time required to read and answer all questions) were removed to produce a final validated data set (overall, 122 [15%] of surveys were excluded). The identifiable details related to drug profiles and the sponsor were masked in the online survey, such that participants were not made aware of the sponsor, name of drug profiles used to derive attributes and levels, or dosage strengths. The sponsor had no access to the identities of the participants. No product, company, or brand names were mentioned in the survey.

Prior to the start of the survey, in a preamble section, respondents were objectively presented with descriptions of various drug benefits and risks, including the way benefit/risk outcomes and data were interpreted and summarized in the context of a placebo-controlled study. The incidence rates for identified risk attributes in the control group were based on completed phase 3 placebo-controlled clinical trials of empagliflozin, sotagliflozin, and pramlintide. Respondents were made aware of the incidence rates of various risk factors during placebo treatment as observed in clinical trials (4,7,8). Patients were also briefed and thereafter tested in the preamble section of the survey about the meaning of a placebo-controlled trial and the interpretation of incidence rates and incidence rate ratios (relative risks).

For the conjoint analysis section of this survey, respondents were presented with six pairwise choices between “packages” of eight attributes (i.e., medication benefits and side effects/risks). Their answers were used to calculate the importance of one attribute relative to another for the entire population and for different segments of the respondent pool. These data were also used to objectively estimate patient drug profile preference outcomes in the overall respondent pool and subgroups.

In addition, in the explicit choice section of the survey, direct respondent preferences, with respondents having full knowledge of the three drug profiles, were calculated, and the respondent likelihood to take the various drug options was also quantified.

The results from the conjoint analysis and the explicit choice questions were also analyzed according to the following subgroups (% of total): country (U.S. [33%], Canada [35%], and Germany [33%]), HbA1c of >7.5% to ≤8% (>58 to ≤64 mmol/mol) (33%) or >8% (>64 mmol/mol) (67%), age (18–29 years [16%], 30–44 years [43%], 45–64 years [34%], and ≥65 years [7%]), BMI (<25 kg/m2 [38%], 25 to <30 kg/m2 [32%], and ≥30 kg/m2 [27%]), DKA history (within the last 3 years [24%] and within the last 1 year [9%]), severe hypoglycemia history (within the last 3 years [51%] and within the last 1 year [30%]), CGM (52%) versus no CGM (48%), pump (65%) versus MDI (35%), and female (58%) versus male (42%).

Statistical Analysis

Statistical significance for all analyses was tested at the 90% confidence level using a two-tailed z test. Differences between importance scores in the conjoint analysis were tested for statistical significance at the 90% confidence level to produce a ranking of importance by the scores (summed to 100%). Attribute importance scores and other results were rounded to whole numbers. Based on a pairwise statistical testing strategy, attributes that were not statistically different from one another were assigned the same rank. Scores were tested in descending order against the highest numerical score. All scores statistically similar to a higher score were “promoted” to the higher rank until a statistically significant difference was found.

Baseline Clinical Characteristics

Valid complete surveys from 701 respondents were received (out of 1,078 dispatched) and analyzed, with 231 (33%) completed in the U.S., 242 (35%) in Canada, and 228 (33%) in Germany (Fig. 1A). Of these, 228 patients (33%) had HbA1c of >7.5% to 8.0% (>58 to 64 mmol/mol) and 473 patients (67%) had HbA1c >8.0% (>64 mmol/mol). Mean ± SD HbA1c for the respondent population was 8.5% ± 0.68%) (69 ± 7.4 mmol/mol) and mean duration of diabetes was 23.1 ± 13.3 years.

Figure 1

A: Flow diagram of survey responses and selection. B: Relative importance and ranking of attributes as determined by conjoint analysis. Relative scores expressed in % as weighted out of 100; attributes depicted with the same color are similarly important. C: Conjoint analysis of patient preference outcomes in the overall respondent population (highlighted in yellow) and predefined subgroups. hypo, hypoglycemia.

Figure 1

A: Flow diagram of survey responses and selection. B: Relative importance and ranking of attributes as determined by conjoint analysis. Relative scores expressed in % as weighted out of 100; attributes depicted with the same color are similarly important. C: Conjoint analysis of patient preference outcomes in the overall respondent population (highlighted in yellow) and predefined subgroups. hypo, hypoglycemia.

Conjoint Analysis Findings

Importance of Attributes

The conjoint analysis weighed (out of 100%) the various attributes by their relative importance (Fig. 1B). Based on the overall responder analysis, the risk of DKA was identified as the most important attribute to patients with a relative weighted score of 23% (z test, P < 0.05). This was followed by similarly ranked HbA1c reduction (14%), risk of severe hypoglycemia (13%), oral versus injection treatment (12%), and risk of genital infection (12%). Risk of nausea was ranked next (11%), followed by weight reduction (8%) and risk of diarrhea (7%), which were ranked least important (Fig. 1B). Subgroup analyses did not show statistically significant differences regarding attribute importance and ranking compared with the rankings obtained in the overall population (data not shown).

Respondent Preference Outcomes

As part of the conjoint analysis and based on the set of attributes used in the analysis, drug profiles comparable to a low-dose oral SGLTi had a significantly higher preference outcome in the overall population relative to the high-dose oral SGLTi profile or the currently available injectable adjunctive therapy profile (see Fig. 1C). The low-dose SGLTi profile garnered 83% of the total preference outcomes for all respondents (z test, P < 0.05) compared with 8% for a high-dose SGLTi option and 9% for the current therapy option (Fig. 1B). Notably, the preference outcome for a low-dose SGLTi option was 95%, 82%, and 74% in the U.S., Germany, and Canada, respectively.

Head-to-head Explicit Medication Profile Comparison

When respondents were asked directly which of the medication profiles (masked to drug name and dose) they would select if they were to start on a medication in addition to their insulin regimen, 69% selected a low-dose oral SGLTi drug profile, 17% selected the high-dose oral SGLTi option, and 6% selected the current injectable adjunctive therapy option, while 9% opted to not take any of the three choices (z test, P < 0.05) (see Fig. 2A).

Figure 2

A: Explicit patient responses for preferred adjunct-to-insulin therapy choice. B: Explicit likelihood-to-adopt responses for the three proposed medication options; the adjusted likelihood score is defined as a weighted score represented by 50% of the “definitely” score plus 25% of the “likely” score.

Figure 2

A: Explicit patient responses for preferred adjunct-to-insulin therapy choice. B: Explicit likelihood-to-adopt responses for the three proposed medication options; the adjusted likelihood score is defined as a weighted score represented by 50% of the “definitely” score plus 25% of the “likely” score.

Estimates of the Likelihood to Adopt Each Medication Profile

When asked about their likelihood to take each medication, respondents were significantly more likely to say they would “definitely” or “likely” take a low-dose SGLTi (35% and 52%, respectively) compared with high-dose SGLTi (7% and 39%, respectively) or currently available T1D adjunctive injection therapy (5% and 40%, respectively) (see Fig. 2B). The adjusted likelihood score (a weighted score represented as 50% of the “definitely” score plus 25% of the “likely” score) was also significantly higher for a low-dose oral SGLTi at 31% vs. 13% for a high-dose oral SGLTi and 12% for currently available adjunctive injectable therapy (P < 0.05).

To our knowledge, this survey with a validated methodology is the first to highlight the relative importance that persons living with T1D place on the various risk and benefit attributes of adjunct-to-insulin therapies for the management of glycemic control in T1D. In the decision-making process to supplement an insulin regimen with another pharmacological agent, these results first indicate that although patients seek glycemic and weight benefits, they are most concerned about safety considerations such as the risk of DKA imposed by that therapy. Second, these results indicate that in comparing risk profiles between a lower- or higher-dose oral SGLTi and a subcutaneous therapy, or the option of maintaining therapy with insulin alone, the balance of risk and benefit shows a clear patient preference for the lesser-risk strategy of low-dose oral SGLTi.

These findings were demonstrated in this survey using different methodologies. The iterative conjoint analysis method showed that 83% of participants prefer the profile of an adjunct-to-insulin therapy such as a low-dose SGLTi, where the risk of DKA is lower relative to the higher-dose SGLTi, and the oral route was preferred rather than the injectable route. The explicit choice method demonstrated that 69% chose a therapy profile similar to a low-dose oral SGLTi compared with the other therapies or insulin alone. The likelihood-to-adopt questionnaire indicated that 87% of participants were likely (52%) or definite (35%) regarding their decision to adopt such a therapy, with proportions much higher than for the other options. Indeed, only a minority (12%) of persons were unlikely or definitely not willing to take a low-dose SGLTi in contrast to the majority who were unlikely or definitely not willing to take high-dose SGLTi (53%) or subcutaneous injection therapy (56%) as adjunctive therapy.

This systematic approach to determining the patient voice in direct response to their review of the data from phase 3 clinical trials represents important information to be considered by patients and clinicians in the decision-making process for selecting new therapies.

Long-term glycemic control data from clinical trial participants (20) and from population-based cohort studies (2,3) demonstrate that only a small proportion of persons with T1D maintain target levels of HbA1c. Therefore, therapeutic choices to supplement intensive insulin therapy would be highly desirable by clinicians and patients. The SGLTi class of medications has a clear effect on improving HbA1c without an increase in rate of hypoglycemia, while showing modest weight reduction and blood pressure lowering, small decreases in insulin dose requirements, and increased time in target glucose range (6,810,12,2124). Additionally, these agents have the potential benefit, based on trials in patients with T2D, to improve cardiovascular and renal outcomes (14,15,2527).

For provision of a clear net benefit to patients, adjunctive-to-insulin therapy options should not increase the risk of adverse events in T1D. Risk minimization of acute complications in T1D is particularly important from a clinical perspective given the current observation that in clinical practice 4.8% of patients experience at least one DKA event per year and 11.8% experience at least one severe hypoglycemic event per year (28), without the use of adjunctive-to-insulin therapies. Systematic use of pramlintide therapy might be expected to increase the baseline rates of severe hypoglycemia, likely in part explaining the very low use in the U.S. despite its indication as an adjunct-to-insulin agent in T1D (5). Similarly, the systematic use of SGLTi, if approved, might be expected to increase the baseline rates of DKA, requiring specific mitigation strategies.

European regulatory approval for use of dapagliflozin in T1D in patients with BMI ≥27 kg/m2 was gained in February 2019 and sotagliflozin in March 2019 with similar restrictions. Ipragliflozin (in December 2018) and dapagliflozin (in March 2019) also gained regulatory approval in Japan for use in T1D. Accordingly, broader clinical use of SGLTi in persons with T1D will require implementation of DKA risk mitigation plans. These should include informative drug labels for safe use of SGLTi in this setting, as well as intensive educational programs including tools such as a medication guide, prescriber-oriented and patient-oriented education and training platforms, medication alert cards, and DKA risk recognition strategies such as the smart use of capillary ketone monitoring (29,30). Finally, the use of lower-dose SGLTi, associated in clinical trials with low risk of DKA comparable to that of placebo (8), should be considered as another component of the DKA mitigation strategy. This is consistent with recent clinical perspectives on potential safe use of SGLTi in T1D in which recommendations are made for the clinical use of the lowest available dose of any efficacious agent in order to lower as much as possible the risk of DKA (1618).

The concept of a low-dose SGLTi strategy also seems to be preferred by patients where the tradeoff for slightly lower glycemic efficacy in terms of HbA1c lowering in exchange for reduced DKA risk is favored. Current results demonstrate objectively that a potential three- to fivefold increased DKA risk is of most concern and highly relevant from the patient perspective, seen as the most important attribute relative to other attributes including the potential for an HbA1c lowering of nearly 0.5% (5.5 mmol/mol). The choice for the most optimal adjunct-to-insulin therapy, from a patient perspective, was therefore impacted to a large degree by the increased risk of DKA. Understanding and reducing the risk of DKA in general is especially important given recent data indicating an increased rate of DKA in the U.S. from 2009 to 2014, since SGLTi was only first approved for use in T2D in 2013 (31).

The results of this scientific survey have several strengths. One is the objective nature of the study utilizing the well-established and validated discrete choice experiment form of conjoint analysis (19,32). The core of conjoint analysis relies on the untangling of the complex tradeoffs patients unconsciously make when selecting one drug product over another, allowing for the objective qualification of patient preferences. Based on the reliability of this unbiased analysis methodology, its use to measure patient preferences has increased in health care research where health authorities such as the U.S. Food and Drug Administration (Center for Devices and Radiological Health) also value its utility (3336). Other strengths of the survey are the clinical applicability and relevance of the data generated, given the baseline clinical characteristics of the selected respondent population. The population studied was on different insulin regimens across a wide range of ages, with an overall glycemic control (mean HbA1c of 8.5% [69 mmol/mol]) that is close to what is currently achievable in real-world care, as evidenced by the latest T1D Exchange data with a mean HbA1c of 8.4% (68 mmol/mol) (2,3).

Limitations include the extent of the selection of attributes included in the conjoint analysis (limited to eight efficacy and safety outcomes in total), that the effect size of some attributes was determined from individual trials rather than a summary measure of multiple trials, and that glucose time in range was not included in the analysis. Use of glucose time in range could not be considered, as it was not evaluated for all therapies (pramlintide) and is not systematically used by all patients. Other limitations of the study include the lack of assessment of preference outcomes in patients close to or at glycemic target (HbA1c <7.5% [<58 mmol/mol]) as well as the lack of inclusion of patients beyond North America and Europe. Despite these limitations, overall, the results of this scientific survey may help to better characterize the efficacy versus safety tradeoff choices and patient preferences in terms of the potential use of adjunct-to-insulin therapy options in T1D. The clear preference of patients for a low-dose oral SGLTi option as an adjunct-to-insulin therapy was quite robust using a validated method for the evaluation of preferences.

The results of this scientific survey are clinically significant at this juncture in time, given recent approvals of some SGLTi as adjunct-to-insulin therapy in T1D (thus far for sotagliflozin and dapagliflozin in Europe) and ongoing efforts elsewhere to gain regulatory approvals. These survey results will be highly informative for potential prescribers of SGLTi in T1D and their dialogue with patients in choosing an optimal and customized therapy option. These results also highlight the importance of the DKA risk with SGLTi use in T1D as a genuine concern for the patients and as a critical topic in the overall benefit-risk analysis of these new agents for T1D. There is a clear need to effectively reduce this important risk in clinical practice (once agents gain regulatory approval) where, in addition to effective DKA education measures, one important strategy to consider would be the use of a lower approved SGLTi dose.

Acknowledgments. The authors sincerely thank Jonathon Gibbs, Certified Medical Publication Professional of Elevate Scientific Solutions, for graphical support (supported financially by Boehringer Ingelheim).

Funding and Duality of Interest. This study was supported by Boehringer Ingelheim and Eli Lilly and Company. B.A.P. has received speaker honoraria from Medtronic, Johnson & Johnson, Dexcom, Insulet, Novo Nordisk, AstraZeneca, Abbott, and Sanofi; has received research grant support from Boehringer Ingelheim, Medtronic, Novo Nordisk, and the Bank of Montreal; and has served as a consultant for Boehringer Ingelheim, Novo Nordisk, Insulet, Sanofi, Abbott, and NeuroMetrix. J.R. has served on scientific advisory boards and received honorarium or consulting fees from Eli Lilly, Novo Nordisk, Sanofi, Janssen, Boehringer Ingelheim, and Intarcia and has received grants/research support from Merck, Pfizer, Sanofi, Novo Nordisk, Bristol-Myers Squibb, Eli Lilly, GlaxoSmithKline, AstraZeneca, Janssen, Genentech, Boehringer Ingelheim, Intarcia, and Lexicon. J.S.S. has acted as an advisor to Adocia, Applied Therapeutics, AstraZeneca, Boehringer Ingelheim, DalCor, Dance Biopharm, DiaVacs, Duologics, Elcelyx Therapeutics, Eli Lilly, Esperion, Geneuro, Ideal Life, Immunomolecular Therapeutics, Intarcia, Intrexon/ActoBio, Kamada, Merck, Orgenesis, Sanofi, Servier, Tolerion, vTv, Valeritas, Viacyte, and Zafgen; has research funding from the National Institutes of Health, JDRF, and the Diabetes Research Institute Foundation; chairs the Strategic Advisory Board of the European Union INNODIA consortium; and has served as a member of the boards of directors of Dexcom, Intarcia, and Moerae Matrix. L.M.L. has been a consultant for Johnson & Johnson, Eli Lilly, Sanofi, Novo Nordisk, MannKind, Merck, Bristol-Myers Squibb, AstraZeneca, Roche, Dexcom, Unomedical-ConvaTec, Insulet, and Boehringer Ingelheim and has received grant support from the National Institutes of Health, JDRF, the American Diabetes Association, the Helmsley Charitable Trust, Dexcom, Insulet, Boehringer Ingelheim, Sanofi, and Novo Nordisk. D.Z.C. has received honoraria from Boehringer Ingelheim, Eli Lilly, Merck, AstraZeneca, Sanofi, Mitsubishi Tanabe Pharma, AbbVie, Janssen, Bayer, and Prometic and has received operational funding for clinical trials from Boehringer Ingelheim, Eli Lilly, Merck, Janssen, Sanofi, and AstraZeneca. C.M. serves or has served on the advisory panels for Novo Nordisk, Sanofi, Merck Sharp & Dohme, Eli Lilly and Company, Novartis, AstraZeneca, Boehringer Ingelheim, Hanmi Pharmaceuticals, Roche, Medtronic, ActoBio Therapeutics, Pfizer, dianax, and UCB Pharma, Inc. Financial compensation for these activities has been received by KU Leuven. KU Leuven has received research support for C.M. from Medtronic, Novo Nordisk, Sanofi, Merck Sharp & Dohme, Eli Lilly, Roche, Abbott, ActoBio Therapeutics, and Novartis, and C.M. serves or has served on the speakers’ bureaus for Novo Nordisk, Sanofi, Merck Sharp & Dohme, Eli Lilly, Boehringer Ingelheim, AstraZeneca, and Novartis. Financial compensation for these activities has been received by KU Leuven. C.P. and R.W. are employees of dQ&A. O.K., J.T.G., J.M., and N.S. are employees of Boehringer Ingelheim. No other potential conflicts of interest relevant to this article were reported.

Author Contributions. B.A.P., J.R., J.S.S., L.M.L., D.Z.C., C.M., C.P., R.W., O.K., J.T.G., J.M., and N.S. took part in the analysis and interpretation of data, as well as the drafting and critical revision of the manuscript; had access to the data included in this publication; and had the ultimate responsibility for the decision to publish this work. B.A.P. and N.S. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Prior Presentation. Parts of this study were presented in abstract form at the 79th Scientific Sessions of the American Diabetes Association, San Francisco, CA, 7–11 June 2019.

1.
Davies
MJ
,
D’Alessio
DA
,
Fradkin
J
, et al
.
Management of hyperglycemia in type 2 diabetes, 2018. A consensus report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD)
.
Diabetes Care
2018
;
41
:
2669
2701
[PubMed]
2.
Foster
NC
,
Beck
RW
,
Miller
KM
, et al
.
State of type 1 diabetes management and outcomes from the T1D Exchange in 2016-2018
.
Diabetes Technol Ther
2019
;
21
:
66
72
[PubMed]
3.
Miller
KM
,
Foster
NC
,
Beck
RW
, et al.;
T1D Exchange Clinic Network
.
Current state of type 1 diabetes treatment in the U.S.: updated data from the T1D Exchange clinic registry
.
Diabetes Care
2015
;
38
:
971
978
[PubMed]
4.
Ratner
RE
,
Dickey
R
,
Fineman
M
, et al
.
Amylin replacement with pramlintide as an adjunct to insulin therapy improves long-term glycaemic and weight control in type 1 diabetes mellitus: a 1-year, randomized controlled trial
.
Diabet Med
2004
;
21
:
1204
1212
[PubMed]
5.
Lyons
SK
,
Hermann
JM
,
Miller
KM
, et al
.
Use of adjuvant pharmacotherapy in type 1 diabetes: international comparison of 49,996 individuals in the Prospective Diabetes Follow-up and T1D Exchange registries
.
Diabetes Care
2017
;
40
:
e139
e140
[PubMed]
6.
Dandona
P
,
Mathieu
C
,
Phillip
M
, et al.;
DEPICT-1 Investigators
.
Efficacy and safety of dapagliflozin in patients with inadequately controlled type 1 diabetes (DEPICT-1): 24 week results from a multicentre, double-blind, phase 3, randomised controlled trial
.
Lancet Diabetes Endocrinol
2017
;
5
:
864
876
[PubMed]
7.
Garg
SK
,
Henry
RR
,
Banks
P
, et al
.
Effects of sotagliflozin added to insulin in patients with type 1 diabetes
.
N Engl J Med
2017
;
377
:
2337
2348
[PubMed]
8.
Rosenstock
J
,
Marquard
J
,
Laffel
LM
, et al
.
Empagliflozin as adjunctive to insulin therapy in type 1 diabetes: the EASE trials
.
Diabetes Care
2018
;
41
:
2560
2569
[PubMed]
9.
Buse
JB
,
Garg
SK
,
Rosenstock
J
, et al
.
Sotagliflozin in combination with optimized insulin therapy in adults with type 1 diabetes: the North American inTandem1 Study
.
Diabetes Care
2018
;
41
:
1970
1980
[PubMed]
10.
Danne
T
,
Cariou
B
,
Banks
P
, et al
.
HbA1c and hypoglycemia reductions at 24 and 52 weeks with sotagliflozin in combination with insulin in adults with type 1 diabetes: the European inTandem2 Study
.
Diabetes Care
2018
;
41
:
1981
1990
[PubMed]
11.
Dandona
P
,
Mathieu
C
,
Phillip
M
, et al.;
DEPICT-1 Investigators
.
Efficacy and safety of dapagliflozin in patients with inadequately controlled type 1 diabetes: The DEPICT-1 52-week study
.
Diabetes Care
2018
;
41
:
2552
2559
[PubMed]
12.
Mathieu
C
,
Dandona
P
,
Gillard
P
, et al.;
DEPICT-2 Investigators
.
Efficacy and safety of dapagliflozin in patients with inadequately controlled type 1 diabetes (the DEPICT-2 study): 24-week results from a randomized controlled trial
.
Diabetes Care
2018
;
41
:
1938
1946
[PubMed]
13.
Wanner
C
,
Inzucchi
SE
,
Lachin
JM
, et al.;
EMPA-REG OUTCOME Investigators
.
Empagliflozin and progression of kidney disease in type 2 diabetes
.
N Engl J Med
2016
;
375
:
323
334
[PubMed]
14.
Zinman
B
,
Wanner
C
,
Lachin
JM
, et al.;
EMPA-REG OUTCOME Investigators
.
Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes
.
N Engl J Med
2015
;
373
:
2117
2128
[PubMed]
15.
Wiviott
SD
,
Raz
I
,
Bonaca
MP
, et al.;
DECLARE–TIMI 58 Investigators
.
Dapagliflozin and cardiovascular outcomes in type 2 diabetes
.
N Engl J Med
2019
;
380
:
347
357
[PubMed]
16.
Nathan
DM
.
Adjunctive treatments for type 1 diabetes
.
N Engl J Med
2017
;
377
:
2390
2391
[PubMed]
17.
Petrie
JR
.
SGLT2 inhibitors in type 1 diabetes: knocked down, but up again
?
Lancet Diabetes Endocrinol
2017
;
5
:
841
843
[PubMed]
18.
McCrimmon
RJ
,
Henry
RR
.
SGLT inhibitor adjunct therapy in type 1 diabetes
.
Diabetologia
2018
;
61
:
2126
2133
[PubMed]
19.
Bridges
JF
,
Hauber
AB
,
Marshall
D
, et al
.
Conjoint analysis applications in health--a checklist: a report of the ISPOR Good Research Practices for Conjoint Analysis Task Force
.
Value Health
2011
;
14
:
403
413
[PubMed]
20.
Nathan
DM
,
Bayless
M
,
Cleary
P
, et al.;
DCCT/EDIC Research Group
.
Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications Study at 30 years: advances and contributions
.
Diabetes
2013
;
62
:
3976
3986
[PubMed]
21.
Chen
J
,
Fan
F
,
Wang
JY
, et al
.
The efficacy and safety of SGLT2 inhibitors for adjunctive treatment of type 1 diabetes: a systematic review and meta-analysis
.
Sci Rep
2017
;
7
:
44128
[PubMed]
22.
El Masri
D
,
Ghosh
S
,
Jaber
LA
.
Safety and efficacy of sodium-glucose cotransporter 2 (SGLT2) inhibitors in type 1 diabetes: a systematic review and meta-analysis
.
Diabetes Res Clin Pract
2018
;
137
:
83
92
[PubMed]
23.
Yamada
T
,
Shojima
N
,
Noma
H
,
Yamauchi
T
,
Kadowaki
T
.
Sodium-glucose co-transporter-2 inhibitors as add-on therapy to insulin for type 1 diabetes mellitus: systematic review and meta-analysis of randomized controlled trials
.
Diabetes Obes Metab
2018
;
20
:
1755
1761
[PubMed]
24.
Yang
XP
,
Lai
D
,
Zhong
XY
,
Shen
HP
,
Huang
YL
.
Efficacy and safety of canagliflozin in subjects with type 2 diabetes: systematic review and meta-analysis
.
Eur J Clin Pharmacol
2014
;
70
:
1149
1158
[PubMed]
25.
Wanner
C
,
Lachin
JM
,
Inzucchi
SE
, et al.;
EMPA-REG OUTCOME Investigators
.
Empagliflozin and clinical outcomes in patients with type 2 diabetes mellitus, established cardiovascular disease, and chronic kidney disease
.
Circulation
2018
;
137
:
119
129
[PubMed]
26.
Cherney
DZ
,
Perkins
BA
,
Soleymanlou
N
, et al
.
Renal hemodynamic effect of sodium-glucose cotransporter 2 inhibition in patients with type 1 diabetes mellitus
.
Circulation
2014
;
129
:
587
597
[PubMed]
27.
Neal
B
,
Perkovic
V
,
Mahaffey
KW
, et al.;
CANVAS Program Collaborative Group
.
Canagliflozin and cardiovascular and renal events in type 2 diabetes
.
N Engl J Med
2017
;
377
:
644
657
[PubMed]
28.
Weinstock
RS
,
Xing
D
,
Maahs
DM
, et al.;
T1D Exchange Clinic Network
.
Severe hypoglycemia and diabetic ketoacidosis in adults with type 1 diabetes: results from the T1D Exchange clinic registry
.
J Clin Endocrinol Metab
2013
;
98
:
3411
3419
[PubMed]
29.
Garg
SK
,
Peters
AL
,
Buse
JB
,
Danne
T
.
Strategy for mitigating DKA risk in patients with type 1 diabetes on adjunctive treatment with SGLT inhibitors: a STICH protocol
.
Diabetes Technol Ther
2018
;
20
:
571
575
[PubMed]
30.
Danne T, Garg S, Peters AL, et al. International consensus on risk management of diabetic ketoacidosis in patients with type 1 diabetes treated with sodium–glucose cotransporter (SGLT) inhibitors. Diabetes Care 2019;42:1147–1154
31.
Benoit
SR
,
Zhang
Y
,
Geiss
LS
,
Gregg
EW
,
Albright
A
.
Trends in diabetic ketoacidosis hospitalizations and in-hospital mortality - United States, 2000-2014
.
MMWR Morb Mortal Wkly Rep
2018
;
67
:
362
365
[PubMed]
32.
Ryan
M
,
Farrar
S
.
Using conjoint analysis to elicit preferences for health care
.
BMJ
2000
;
320
:
1530
1533
[PubMed]
33.
Hofheinz
R
,
Clouth
J
,
Borchardt-Wagner
J
, et al
.
Patient preferences for palliative treatment of locally advanced or metastatic gastric cancer and adenocarcinoma of the gastroesophageal junction: a choice-based conjoint analysis study from Germany
.
BMC Cancer
2016
;
16
:
937
[PubMed]
34.
Johnson
FR
,
Zhou
M
.
Patient preferences in regulatory benefit-risk assessments: a US perspective
.
Value Health
2016
;
19
:
741
745
[PubMed]
35.
Louder
AM
,
Singh
A
,
Saverno
K
, et al
.
Patient preferences regarding rheumatoid arthritis therapies: a conjoint analysis
.
Am Health Drug Benefits
2016
;
9
:
84
93
[PubMed]
36.
Mansfield
C
,
Sikirica
MV
,
Pugh
A
, et al
.
Patient preferences for attributes of type 2 diabetes mellitus medications in Germany and Spain: an online discrete-choice experiment survey
.
Diabetes Ther
2017
;
8
:
1365
1378
[PubMed]
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.