This register-based retrospective cohort study aimed to estimate the consequences of a wide-reaching implementation of the SGLT2 inhibitor empagliflozin for patients with T2D and established CVD. Though there are other novel glucose-lowering medications, this study was limited to empagliflozin, as it is the recommended treatment according to regional (Östergötland) and national (Sweden) guidelines. Risk reductions regarding cardiovascular events and mortality as identified in the EMPA-REG OUTCOME study were applied on a geographically defined cohort of patients with T2D and established CVD, in the region of Östergötland, Sweden (the T2D CVD cohort). At baseline, i.e. during the first 3 months of 2012, 85% of the patients in the T2D CVD cohort received antihypertensive treatment and 58% received lipid-lowering treatment, indicating that there was room for improvement according to the national guidelines. The effects of applying the EMPA-REG OUTCOME study risk reductions to all patients with T2D and established CVD during a 5-year period included reduced healthcare utilization and improved survival, at a low cost. The results suggest that utilization of empagliflozin for patients with T2D and established CVD would be a good use of healthcare resources. The estimated decrease in the costs of healthcare visits was approximately SEK 4100; increasing drug costs (for empagliflozin) resulted in a net cost increase of approximately SEK 18,000. This increase should be evaluated in the context of the survival gain of 96 days per treated patient, over a period of 5 years.
The results from this study indicate that the increased empagliflozin utilization would cost approximately SEK 69,000 per life-year gained. Applying a quality of life decrement of 0.19 (on a scale 0–1) for patients with T2D and established CVD  the result would translate to a cost per quality-adjusted life-year (QALY) of just over SEK 85,000. These preliminary results should be interpreted with caution as we have not applied a lifetime perspective on survival and costs. Our findings, however, indicate that empagliflozin is cost-effective, as the estimated cost per QALY is well below both commonly applied threshold values  and the marginal productivity of the healthcare sector .
For the public healthcare region of Östergötland, 5 years of treatment with empagliflozin for the 5490 patients with T2D and established CVD identified in this study would result in increased costs of approximately SEK 100 million (SEK 20 million per year), and 1455 life-years gained (approximately 290 life-years per year of treatment). Extrapolated to the country level, each year of treatment with empagliflozin would result in approximately 5800 life-years gained at a cost of approximately SEK 440 million.
Our results indicate that treating this vulnerable patient population with empagliflozin has positive effects on health and is also cost-effective. Treatment of this patient group should follow international and national guidelines [2, 5]. Previous research shows that the implementation and uptake of new diabetes medications in Sweden is generally slow, placing it in this respect far behind its neighbouring countries .
The main strength of the study is that it is based on patient data from high-quality registers, which means that the mapping of the cohort from Östergötland is very reliable. Information regarding healthcare utilization, drug consumption, costs and mortality of this patient cohort from 2012 to 2016 are well recorded in these registers.
Previous research has shown that glucose-lowering drugs for patients with T2D and established CVD have a beneficial effect with a reduced risk of cardiovascular events [7, 8, 17]. The beneficial effects are particularly noticeable for SGLT2 inhibitors . Costs associated with the treatment of patients with T2D and established CVD have been studied to a lesser extent. A study conducted in the USA  compared two different glucose-lowering drugs, one SGLT2 inhibitor (dapagliflozin) and the other a dipeptidyl peptidase 4 inhibitor (sitagliptin). It was concluded that dapagliflozin led to higher drug costs but that this was offset by lower healthcare costs. In a Swedish registry-based study , dapagliflozin was compared with “other glucose-lowering drugs” (other SGLT2 inhibitors excluded). This study also concluded that dapagliflozin reduced hospital care costs related to CVD, while also leading to higher drug costs for patients with T2D and established CVD or multiple risk factors for CVD. Comparing different SGLT2 inhibitors, it has been suggested that empagliflozin prescribed for preventing CV death and hospitalisation for heart failure in patients with T2D is cost-effective compared to treatment with canagliflozin or dapagliflozin .
Our results are associated with some uncertainty. Even though the inclusion criteria of the EMPA-REG OUTCOME study were applied when including patients in the T2D CVD cohort in Östergötland, some differences between the populations were apparent. For example, the Östergötland T2D CVD cohort was older than the patients in the EMPA-REG OUTCOME study and contained a greater proportion of women. Although the CVD cohort in Östergötland was older, it included a lower proportion of patients with long (> 10 years) diabetes duration and, on average, patients had lower HbA1c values compared to the EMPA-REG OUTCOME study population. However, the Östergötland patients had poorer kidney function, which is reasonable given the age difference. As a result of these differences the study should be regarded as exploratory only.
Caution should be exercised when it comes to the generalizability of results reported in this study. Results are specific to patients with T2D and established CVD and should not be extrapolated to patients with T2D without established CVD (primary prevention). As the study was conducted in Sweden, there may be differences in this patient population that would make generalizing the results to other countries problematic, such as differences in patient population characteristics, national treatment guidelines and therapeutic treatment in T2D.