Introduction

In people with type 2 diabetes (T2D), atherosclerotic cardiovascular disease (ASCVD) is the leading cause of morbidity and mortality and is associated with significant health care utilization and cost [1,2,3,4]. Several large cardiovascular outcome trials (CVOTs) have demonstrated that glucose-lowering agents belonging to the glucagon-like peptide-1 receptor agonist (GLP-1 RA) class significantly reduce cardiovascular events vs baseline glucose-lowering therapy in high-risk individuals with T2D [1, 5, 6]. In keeping with these data, diabetes and cardiology guidelines and professional societies recommend a GLP-1 RA with demonstrated cardiovascular benefit for the treatment of individuals with T2D and established or at high risk for ASCVD. This recommendation is independent of baseline or target glycated hemoglobin (HbA1c) levels [1, 5,6,7].

Dipeptidyl peptidase-4 inhibitors (DPP-4is) are also incretin-based therapies commonly used in clinical practice for people with T2D [8]. While GLP-1 RAs have demonstrated cardiovascular benefits in CVOTs, DPP-4is have not [8]. However, head-to-head trials specifically comparing the effects of these 2 drug classes on cardiovascular outcomes in individuals with T2D and established ASCVD are lacking. It is important to evaluate their cardiovascular benefits in longitudinal, real-world studies to explore the effects of these therapies in larger populations who do not take part in clinical trials and to analyze health care utilization and costs in these populations.

Some evidence has also suggested that the cardiovascular benefits of GLP-1 RAs may not be a class effect [8]. Older generations of daily GLP-1 RAs may be less efficacious in reducing cardiovascular events [8]; therefore, including them in assessments of the entire GLP-1 RA class may generate mixed effects. It is instead important to examine the cardiovascular effectiveness of the newer, once-weekly (OW) generation of GLP-1 RAs, which have been linked to improvements in glycemic control [9, 10], reduced body weight [9, 11], and better adherence/persistence relative to daily injectable GLP-1 RAs [12] and now comprise the vast majority of prescriptions for this class in the US [13]. However, real-world evidence comparing cardiovascular outcomes in US adults with T2D and established ASCVD who initiated a OW GLP-1 RA vs a DPP-4i is limited.

This study aims to compare the time to occurrence of ischemic stroke, MI, and their composite, and ASCVD-related and all-cause health care resource utilization (HCRU) and medical costs in people with T2D and ASCVD who initiated a OW GLP-1RA vs a DPP-4i.

Methods

Study design

This was an observational cohort study using the Optum Clinformatics® Data Mart (CDM) database to compare cardiovascular outcomes in adults with T2D and established ASCVD who initiated a OW GLP-1 RA or a DPP-4i. The study period was from January 1, 2017, to September 30, 2021. ASCVD history was examined back to 2001. OW GLP-1 RAs studied included exenatide, dulaglutide, and semaglutide. DPP-4is included sitagliptin, saxagliptin, linagliptin, and alogliptin. Individuals who switched within the same drug class during the follow-up period were included, but those who switched between classes were excluded.

The index date, which was between January 1, 2018, and June 30, 2021, was defined as the prescription date of the index drug (OW GLP-1 RA or DPP-4i). The beginning of the index window was selected to fall immediately after the approval of the newest agent included in this study (once weekly semaglutide, approved for T2D in December 2017). One year prior to the index date constituted the baseline period. Individuals were followed for ≥ 3 months, until censoring due to the earliest of death; end of the study (September 30, 2021); new initiation of a sodium-glucose cotransporter-2 (SGLT-2) inhibitor, DPP-4i (among those in the GLP-1 RA group), or GLP-1 RA (among those in the DPP-4i group); lapse of continuous enrollment; or discontinuation of the index drug for > 60 days. The follow-up period, in which outcomes/end points were assessed, was the interval between the index date and end of follow-up.

Data source

The Optum CDM contains administrative claims for enrollees of commercial health care plans and Medicare Advantage across the US. The administrative claims include verified, adjudicated, adjusted, and de-identified medical and pharmacy claims. The CDM also includes outpatient laboratory test results from large national laboratory vendors.

Study population

Individuals included in the study had ≥ 2 separate diagnoses of T2D on different dates during the study period, identified using the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) code E11 in the primary or secondary diagnosis positions; ≥ 1 prescription for the index drug; and use of the index drug for ≥ 90 days (with ≤ 60-day gaps). Individuals were ≥ 18 years old on the index date and had a history of ASCVD between 2001 and the index date, identified using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and ICD-10-CM codes in any position (see Additional file 1).

Exclusion criteria included baseline GLP-1 RA or DPP-4i use, missing demographic information (age or sex), or pregnancy or type 1 diabetes at any time during the baseline or follow-up periods.

Detailed baseline characteristics are listed in Table 1 and Additional file 2. As this study used only de-identified patient records and did not involve the collection, use or transmittal of individually identifiable data, institutional review board approval was not required.

Table 1 Selected unweighted and weighted key baseline characteristics among adults with T2D and ASCVD
Full size table

Measurement of outcomes and variables

Clinical effectiveness outcomes included ischemic stroke, MI, and their composite. Ischemic stroke events were identified as a primary diagnosis of inpatient claims using ICD-10-CM code I63 (except I63.1, I63.4, and I63.6; see Additional file 1). One hospitalization for ischemic stroke was considered one stroke event. MI was measured as a primary diagnosis of inpatient claims using ICD-10-CM code I21 or I22 (see Additional file 1). One hospitalization for MI was considered one MI event. An ischemic stroke or MI event was defined as the composite of ischemic stroke and MI [14]. Incidence rates and time to event occurrence were evaluated (events could be incident or recurrent).

HCRU outcomes included ASCVD-related and all-cause outpatient visits, hospitalizations, and emergency room (ER) visits. Cost outcomes included ASCVD-related and all-cause hospitalization costs and total medical costs. Total medical costs included medical costs from outpatient visits, hospitalizations, and ER visits. ASCVD-related HCRU and cost outcomes were measured using ICD-10-CM codes in the primary or secondary position (see Additional file 1). All-cause HCRU and medical costs included costs for any diagnosis, including ASCVD. The Optum CDM reports an estimated cost that is standardized based on a resource-based relative value scale derived from observed costs paid by the insurer, rather than the original paid amount.

Statistical methods

Descriptive statistics were presented for all outcomes and covariates in the OW GLP-1 RA and DPP-4i groups. Counts and frequencies were used for categorical variables and means and standard deviations for continuous variables. Costs were reported as per person per month (PPPM) and adjusted to the year 2021. Incidence of first (or recurrent) stroke or MI in the follow-up (for those without and with a prior history of stroke/MI, respectively) was reported as number of events per 1000 person-years.

To reduce the observed selection bias between the 2 groups, inverse probability of treatment weighting (IPTW) [15] using average treatment effect weights was derived by conducting a logistic regression with the following variables: age, sex, race/ethnicity, geographic region, index year, Charlson Comorbidity Index, diabetes complications severity index, type of ASCVD history, comorbidities, glucose-lowering therapy, HbA1c, body mass index (BMI), and number of all-cause hospitalizations 60 days prior to the index date. For baseline characteristics, descriptive statistics were reported with and without IPTW. Standardized mean differences (SMDs) were presented. SMD < 10% was considered not significantly different for baseline characteristics. Weighted descriptive statistics were reported for all outcomes. For time-to-event outcomes, weighted cumulative incidence curves were generated and log-rank tests and Cox proportional-hazards (Cox-PH) regressions were conducted (proportional-hazards assumptions were met in these models). For HCRU and costs, generalized linear models with quasi-Poisson distribution and log-link function were used.

Interaction and stratified analyses were also conducted for the following variables: with and without history of ischemic stroke or MI, and end of follow-up before and after March 1, 2020, for HCRU and cost outcomes (to test the impact of COVID-19 on HCRU and costs). Sensitivity analyses excluding exenatide OW or including the prescriber type in weighting were also conducted. To assess residual unmeasured bias, negative control outcomes (ie, breast cancer and prostate cancer) and E-values were also examined [16]. To reduce bias due to informative censoring, inverse probability of censoring weighting (IPCW) was applied to assess clinical outcomes. To allow for a longer follow-up time, additional sensitivity analyses were conducted with the index date selection window restricted to between 2018 and 2020. Finally, to provide a complementary perspective, intention-to-treat (ITT) analyses were also performed for clinical outcomes.

Results

Before weighting, the study included 26,430 OW GLP-1 RA users and 39,858 DPP-4i users (Additional file 3). After weighting, the sample sizes were 25,287 and 39,684, respectively. The average follow-up time was similar between the two groups (11.3 ± 9.3 months for OW GLP-1 RA; 11.3 ± 8.7 months for DPP-4i). After IPTW weighting, there were no significant differences in baseline characteristics, including use of other glucose-lowering treatment regimens, between the OW GLP-1 RA and DPP-4i groups (Additional file 2). The overall population had a mean age of 70 years and was nearly evenly split between males and females. Approximately 4 in 5 individuals were Medicare Advantage enrollees. Approximately 15% of people had a history of ischemic stroke and 15% had a history of MI. On average, individuals had 4.1 and 5.4 years of ASCVD and T2D history, respectively. In the OW GLP-1 RA group, most patients (62.9%) were taking dulaglutide while the remaining patients were taking once-weekly semaglutide (26.5%) or once-weekly exenatide (10.6%). In the DPP-4i group, most patients (67.3%) were on sitagliptin while the remaining patients were on linagliptin (27.1%), saxagliptin (4.8%), or alogliptin (0.8%).

Clinical effectiveness outcomes

Figure 1 displays the weighted incidence rates, hazard ratios (HRs), and cumulative incidence curves for clinical outcomes. Cox-PH regression results showed that compared with DPP-4is, OW GLP-1 RAs were associated with a 26% lower risk of ischemic stroke (HR [95% CI] 0.74 [0.63–0.87]). Additionally, OW GLP-1 RA users had a 22% lower risk of MI (HR [95% CI] 0.78 [0.67–0.92]) and a 24% lower risk of the composite of ischemic stroke or MI (HR [95% CI] 0.76 [0.68–0.86]) than did DPP-4i users. Incidence rates for stroke, MI, and their composite were − 4.91, − 3.81, and − 8.42 per 1000 person-years lower, respectively, in the OW GLP-1 RA group compared with the DPP-4i group. Similar trends were observed in the cumulative incidence curves for these clinical outcomes (Fig. 1).

Fig. 1
figure 1

Clinical outcomes in adults with T2D and ASCVD on OW GLP-1 RAs or DPP-4is. Weighted incidence rates, hazard ratios, and cumulative incidence curves of clinical outcomes comparing OW GLP-1 RAs with DPP-4is among adults with T2D and ASCVD. *Indicates statistical significance (P < 0.05). ASCVD atherosclerotic cardiovascular disease, DPP-4is, dipeptidyl peptidase-4 inhibitors, GLP-1 RAs glucagon-like peptide-1 receptor agonists, OW once-weekly, T2D type 2 diabetes

Full size image

Health care resource utilization and cost outcomes

Figure 2 depicts the weighted HCRU descriptive statistics and rate ratios (RRs) in the OW GLP-1 RA group vs the DPP-4i group. OW GLP-1 RA users had fewer ASCVD-related and all-cause hospitalizations, ER visits, and outpatient visits. Regression models revealed significant differences in ASCVD-related hospitalizations and outpatient visits and all-cause hospitalizations, ER visits, and outpatient visits. Specifically, compared with DPP-4is, OW GLP-1 RAs were associated with 23% and 13% fewer ASCVD-related hospitalizations and outpatient visits, respectively. OW GLP-1 RA users also had 27%, 8%, and 8% fewer all-cause hospitalizations, ER visits, and outpatient visits, respectively, than did DPP-4i users.

Fig. 2
figure 2

HCRU in adults with T2D and ASCVD on OW GLP-1 RAs or DPP-4is. Weighted descriptive statistics and rate ratios of HCRU comparing OW GLP-1 RAs with DPP-4is among adults with T2D and ASCVD. *Indicates statistical significance (P < 0.05). ASCVD atherosclerotic cardiovascular disease, DPP-4is dipeptidyl peptidase-4 inhibitors, ER emergency room, GLP-1 RAs glucagon-like peptide-1 receptor agonists, HCRU health care resource utilization, IP inpatient, OP outpatient, OW once-weekly, T2D type 2 diabetes

Full size image

Figure 3 shows the weighted descriptive statistics and RRs for medical costs in the OW GLP-1 RA and DPP-4i groups. Compared with DPP-4i users, OW GLP-1 RA users had 25% lower ASCVD-related hospitalization costs (RR [95% CI] 0.75 [0.68–0.84]) and 19% lower ASCVD-related total medical costs (RR [95% CI] 0.81 [0.74–0.88]). Average ASCVD-related hospitalization costs were $341 PPPM (OW GLP-1 RA) vs $453 PPPM (DPP-4i; absolute difference [95% CI] − $111 [− $153 to − $70] PPPM), and average ASCVD-related total medical costs were $597 PPPM (OW GLP-1 RA) vs $738 PPPM (DPP-4i; absolute difference [95% CI] − $141 [− $197 to − $85] PPPM). Compared with DPP-4is, OW GLP-1 RAs were associated with 26% lower all-cause hospitalization costs (RR [95% CI] 0.74 [0.69–0.79]) and 23% lower all-cause total medical costs (RR [95% CI] 0.77 [0.74–0.81]). All-cause hospitalization costs were $789 PPPM (OW GLP-1 RA) vs $1068 PPPM (DPP-4i; absolute difference [95% CI] − $279 [− $341 to − $217] PPPM), and all-cause total medical costs were $2186 PPPM (OW GLP-1 RA) vs $2824 PPPM (DPP-4i; absolute difference [95% CI] − $638 [− $753 to − $523] PPPM).

Fig. 3
figure 3

Medical costs in adults with T2D and ASCVD on OW GLP-1 RAs or DPP-4is. Weighted descriptive statistics and rate ratios of medical costs comparing OW GLP-1 RAs with DPP-4is among adults with T2D and ASCVD. *Indicates statistical significance (P < 0.05). ASCVD atherosclerotic cardiovascular disease, DPP-4is dipeptidyl peptidase-4 inhibitors, GLP-1 RAs glucagon-like peptide-1 receptor agonists, IP inpatient, OW once-weekly, T2D type 2 diabetes

Full size image

Stratified and sensitivity analyses

After excluding exenatide, additional clinical risk reductions were identified in the OW GLP-1 RA group vs the DPP-4i group (Additional file 4). Similar trends were observed for HCRU and cost outcomes. Specifically, OW GLP-1 RAs (dulaglutide and semaglutide OW) were associated with 11% fewer ASCVD-related ER visits vs DPP-4is (HR [95% CI] 0.89 [0.80–0.99]; P = 0.032) (Additional file 5).

Results of sensitivity analyses including the prescriber type in IPTW were very similar to those of the main analyses (Additional file 6 and Additional file 7). The only major difference was that, after balancing on prescriber type, a significant reduction in ASCVD-related ER visits was identified in the OW GLP-1 RA group compared with the DPP-4i group (HR [95% CI] 0.90 [0.81–1.00]; P = 0.041).

No significant interactions were identified between stroke history and time to occurrence of stroke (P = 0.601) or between MI history and time to occurrence of MI (P = 0.794) between the two groups. There were trends toward decreased HCRU and costs in both groups among those who had an overlap with the COVID-19 pandemic compared with those who did not. However, no significant interactions were identified for HCRU and cost outcomes, except for a marginally significant interaction with all-cause hospitalizations (P = 0.050; data not shown). Analyses of negative control outcomes (ie, breast cancer and prostate cancer) showed no significant association. The E-value associated with the stroke outcome, representing the minimal strength of association an unmeasured confounder must have with stroke in the follow-up and treatment selection to change the result of the study, is 2.07 (upper confidence interval: 1.58).

Results generated from an analysis restricting the index date selection window to between 2018 and 2020 mirror the results from the main analysis (Additional file 8). OW GLP-1 RAs were significantly associated with a 27%, 20%, and 23% lower risk of ischemic stroke, MI, and their composite, respectively, compared to DPP-4is (all P < 0.05). The overall conclusions from the IPCW analysis are also similar to those of the main analysis (Additional file 9). With potentially informative censoring taken into consideration, IPCW analysis showed larger risk reductions for all clinical outcomes. More specifically, the IPCW analysis revealed a 31% reduction in risk of stroke, a 27% reduction in risk of MI, and a 29% reduction in risk of their composite in OW GLP-1 RA users compared with DPP-4i users (all P < 0.05). ITT analyses revealed smaller treatment differences between the two groups as compared to the main, per-protocol analysis (Additional file 10), which may be primarily driven by the 44.4% of patients who were not persistent with the index drug.

Discussion

In a large sample of US adults with T2D and established ASCVD, treatment with OW GLP-1 RAs was associated with decreased risk of ischemic stroke and MI and fewer ASCVD-related and all-cause HCRU and medical costs compared with DPP-4is. The samples investigated were adequately sized to detect differences between groups, and individuals were followed for a relatively long time. The study was also designed and conducted to alleviate potential confounding and bias.

Several large, randomized trials have compared cardiovascular outcomes in people treated with OW GLP-1 RAs vs placebo. In the SUSTAIN 6 trial (semaglutide vs placebo), non-fatal stroke occurred in significantly fewer individuals in the semaglutide group (HR 0.61) [17]. Similar findings were observed in the REWIND trial (dulaglutide vs placebo) for both non-fatal stroke (HR 0.76) and non-fatal MI (HR 0.96) [18, 19]. The EXSCEL trial showed directionally similar efficacy results for exenatide OW [20, 21]. A recent meta-analysis of all GLP-1 RAs (including OW and once-daily versions) [22] assessing outcomes across CVOTs found that GLP-1 RAs reduced 3-point MACE by 14%, including a 10% reduction in MI risk and a 17% reduction in stroke risk. Upon exclusion of data from the ELIXA trial, which was conducted in a post-acute coronary syndrome population and used lixisenatide, known to be the least effective GLP-1 RA, risk reductions improved for all outcomes [22]. The HRs reported for stroke and MI in the present study are marginally lower than those reported in the above meta-analysis. Additionally, the sensitivity analysis excluding exenatide OW (which may have a limited effect on the cardiovascular outcomes of interest [20]) revealed lower HRs than did our original analyses including all three GLP-1 RA agents. Of note, the design of this study differed in that it compared OW GLP-1 RAs with DPP-4is and included some population differences (eg, including adults of all ages with T2D and ASCVD).

In real-world studies, GLP-1 RAs have consistently been associated with reduced cardiovascular risk in people with T2D [23,24,25,26,27,28], including reduced incidence of stroke (HRs between 0.65 and 0.74) and MI (HRs between 0.63 and 0.86). However, it is important to note that these studies were conducted in different countries and populations from this study. While previous studies have assessed cardiovascular outcomes in people with T2D and have largely assessed the older generation of GLP-1 RAs, the present study is among the first to examine cardiovascular outcomes in a large US population with T2D and ASCVD and those treated with the newer generation of OW GLP-1 RAs (eg, dulaglutide, semaglutide). However, as shown in clinical trials and in this study, these GLP-1 RAs may not be equivalent in terms of cardiovascular outcomes.

The mechanisms underlying the cardiovascular benefits of GLP-1 RAs are not fully known. However, these mechanisms may relate to a combination of effects on weight, blood pressure, glucose and lipid levels, inflammation, and potential effects on other factors promoting atherothrombosis [5].

Despite guideline recommendations and consistent evidence of cardiovascular risk reduction (including from the present study), DPP-4i use remains prevalent while the adoption of GLP-1 RAs is suboptimal among people with T2D and ASCVD [29], implying the need for a change in clinical practice. It is important that clinicians involved in the care of those with T2D and ASCVD consider or advise the use of GLP-1 RAs as well as other agents proven to lower cardiovascular risk (eg, SGLT-2 inhibitors, antihypertensive agents, lipid-lowering drugs, and antiplatelet agents where indicated) to reduce future atherothrombotic events. Compelling real-world evidence, such as the data provided in this study, supports clinical trial results and provides further evidence that could promote greater adoption of GLP-1 RAs into practice. In this study, the effect size identified for reduction in stroke risk (HR 0.74) is comparable to or exceeds that of meta-analyses investigating the effects of lowering blood pressure (RR 0.73), treating hyperlipidemia (RR 0.79), and other interventions in people with T2D [30], highlighting the benefits that can be achieved by adding GLP-1 RAs to antihyperglycemic regimens.

This study has several limitations. As an observational study, it is limited to assessments of associations and not causality. Additionally, the database used in this study includes predominantly commercial and Medicare Advantage plan enrollees, therefore caution is needed when applying the conclusions from this study to other populations. Finally, potential measurement errors (eg, misclassified billing codes), missing data, unavailability of information of date and cause of death, and use of imputed average cost data (rather than actual paid costs) may further limit the interpretation or applicability of these findings.

Conclusion

Results from a large US administrative claims database revealed that, compared with the use of a DPP-4i, OW GLP-1 RA use among people with T2D and established ASCVD is associated with 26% lower ischemic stroke risk and 22% lower MI risk, along with significant reductions in ASCVD-related hospitalizations and outpatient visits; all-cause hospitalizations, ER visits, and outpatient visits; and ASCVD-related and all-cause hospitalization costs and total medical costs. This real-world evidence supports the findings from CVOTs and guideline recommendations [1, 6, 7] involving this class of glucose-lowering medications.