Type 2 diabetes (T2D) now affects approximately 300,000 people in Aotearoa New Zealand, of whom at least two-thirds will likely die from cardiovascular disease (CVD) and/or renal disease (CVRD).[[1]] Unfortunately, T2D creates some of the greatest inequities for Māori and Pacific peoples[[2,3]] and the shorter life expectancy of Māori and Pacific peoples with T2D in New Zealand has not improved in the past 20 years.[[4]] Addressing the modifiable risk factors responsible, including improving access to healthcare and medications that slow progression or prevent CVRD, is critical in eliminating these disparities.[[4]] This includes removing the inequities in prescribing metformin and statins to prevent CVD and angiotensin-converting enzyme inhibitors (ACEi) or angiotensin receptor blockers (ARB) to slow the progression of renal disease.[[5,6]]

The greatest change to T2D pharmacotherapeutic care in the last two decades has been the use of sodium-glucose co-transporter 2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP1RA). In addition to effectively lowering glucose levels without the risk of hypoglycaemia, these medications lead to weight loss and have actions beyond glycaemic control that reduce the progression of CVRD in people with T2D.[[7]] Hence these medications have the potential to reduce the burden and inequities in CVRD when used in addition to existing medications. Contemporary international and national guidelines recommend using SGLT2i and/or GLP1RA in all with T2D and high risk of CVRD regardless of their glycaemic control.[[8,9]] These guidelines also recommend using SGLT2i and/or GLP1RA in those without CVRD if the HbA1c is >53 mmol/mol despite the use of metformin with or without other glucose lowering therapies.[[8,9]]

After more than a decade of SGLT2i and GLP1RA being the mainstay of treatment of T2D worldwide, PHARMAC finally funded access in New Zealand to empagliflozin (a SGLT2i), from February 2021 and dulaglutide (a GLP1RA), from September 2021.[[10]] Funded access was initially proposed to be restricted under special authority criteria (SAC) for those with T2D with an HbA1c >53mmol/mol despite regular use of at least one glucose lowering therapy for at least 3 months, and either renal disease (eGFR <60mL/min and/or urinary albumin:creatine ratio >3mg/mmol), or CVD, or a 5-year risk of a CV event of >15%. Despite strong evidence of the additive benefits of SGLT2i and GLP1RA on reducing CVRD, funding was limited to either empagliflozin or dulaglutide and not combined use. Although the SAC allowed funded access for one of these agents for high-risk patients with CVRD, there was significant concern that the proposed SAC may widen inequities in access for Māori and Pacific peoples.[[11]] Hence, after significant lobbying from several areas and further consultation, Māori and/or Pacific ethnicity was added as a specific criterion as an alternative to CVRD. Thereby theoretically improving access for Māori and Pacific peoples without CVRD and/or without the need to assess CVRD risk.

Here, we evaluate the impact of including ethnicity as a specific criterion of the SAC by assessing the prescribing of SGLT2i/GLP1RA by Māori, Pacific and non-Māori/non-Pacific (nMnP) ethnicity in those with T2D with CVRD (including equivalent 5-year CV risk of ≥15%), without CVRD (non-CVRD) and unknown CVRD status. Comparisons are made to prescribing unrestricted therapies such as metformin, ACEi/ARBs and statins as per national guidelines.[[8]] Metformin was chosen as the only glucose-lowering therapy comparison as it is the only glucose-lowering therapy recommended for all people with T2D regardless of CVRD status.[[8]] Indications for use in CVRD were taken to be: SGLT2i/GLP1Ra regardless of glycaemic control, metformin if the HbA1c was >45mmol/mol; ACEi or ARB in those with renal disease (defined as per in SAC) and a statin in those with a serum LDL cholesterol (LDLc) >1.8mmol/L.[[8]] Indications in those without CVRD were taken to be: SGLT2i/GLP1RA with HbA1c >53mmol/mol despite the use of at least one glucose lowering therapy and metformin if the HbA1c is >45mmol/mol. Statins and ACEi/ARBs were not used for comparisons in non-CVRD given their use is not essential in this group.

Data were obtained from all ProCare, Hauraki, Pinnacle and National Hauora Coalition practices (n=302) for all enrolled patients with a diagnosis of T2D aged 18–75 years (n=53,142; Table 1) including ethnicity, CV risk, laboratory and prescribing data from  1February 2021 to 31 July 2022). Māori (first) and Pacific ethnicities (second) were prioritised when multiple ethnicities were recorded.

CVRD was defined as the presence of renal disease, CVD or high CV risk as per the SAC (n=24,367). Non-CVRD was defined as those with a normal UACR, eGFR >60mL/min, no documented CVD and a 5-year CV risk <15% (n=12,692). Unknown CVRD was defined when data on their CVRD status were not available (n=16,137). Access to medications (SGLT2i/GLP1RA, metformin, statins and ACEi/ARBs) was defined as the proportion of each ethnicity (Māori, Pacific and nMnP) who received >1 script for each medication in the 18-month study period (numerator) from those clinically recommended for their use (denominator). Patients with an eGFR <15mL/min were excluded from analyses of metformin and SGLT2i/GLP1RA in those with CVRD (n=192), given that their use is not recommended in significant renal impairment. Patients with no HbA1c data in the study period were excluded from analyses in those for metformin with CVRD (n=576), and all analyses for non-CVRD (n=8) and unknown CVRD (n=3,297) given the indication for use is glycaemic control above target. Patients treated with metformin with an HbA1c <45mmol/mol (n=1,464 in CVRD, n=1,061 in non-CVRD and n=1035 in unknown CVRD), statins with an LDLc <1.8mmol/mol (n=207 in CVRD) or SGLTR2i/GLP1RA with an HbA1c <53mmol/mol (n=913 in CVRD, n=3 in non-CVRD and n=277 in unknown CVRD) were included in the numerator for each analysis on the presumption that they met recommendations for use at a timepoint prior to the start of the study period. Patients on non-funded SGLT2i/GLP 1RA alone i.e., dapagliflozin and liraglutide were excluded (n=61). Data were analysed in R version 4.2 with significance accepted at P<0.05.

Here we show that a much greater proportion of Māori and Pacific peoples with CVRD (~12% more), without CVRD (~20% more) or unknown CVRD status (~15% more) have been prescribed SGLT2i/GLP1RA than their nMnP peers (all P<0.05; see Figure 1). In contrast, Māori were prescribed less unrestricted therapies such as statins and ACEi/ARBs in CVRD (~2–3%), and metformin in unknown CVRD than nMnP peoples (all P<0.05). Pacific peoples with CVRD had greater prescribing for all studied unrestricted therapies than Māori and nMnP peoples, but lower rates of prescribed SGLT2i/GLP1RA than Māori when CVRD was absent or unknown (all P<0.05).

View Figures and tables.

The addition of ethnicity as a criterion for funded access to SGLT2i/GLP1RA appears to have been a useful mechanism in addressing the inequity of access seen with unrestricted therapies. Given this study only assessed initiation of therapy and not dispensing, adherence or tolerance of therapy or its effect on outcomes, further research is needed to assess the impact of the SAC. The effect on outcomes is important because it will likely be at least several years before any impact of greater prescribing of SGLT2i/GLP1RA on current inequitable outcomes such as CV events, dialysis, and premature death can be determined. Moreover, our findings show that further actions are required to eliminate inequities for those with T2D and CVRD. Indeed, it is concerning that approximately one in four people with T2D and CVRD are not receiving either metformin, statin or ACEi/ARB therapy, approximately two-thirds are yet to be prescribed SGLT2i/GLP1RA, and dual SGLT2i/GLP1RA therapy remains unfunded. With ongoing workforce and COVID-19 pressures affecting the health system, changes in models of care are likely needed alongside targeted interventions such as education and data dashboards to prioritise and assist prescribers (including nurse and pharmacist prescribers) in improving care in those with T2D and CVRD.

Conclusions

The addition of ethnicity as a criterion to the SAC for SGLT2i and GLP1RA appears to have increased access for Māori and Pacific peoples, but more work is urgently required to eliminate longstanding inequities in CVRD for Māori and Pacific peoples with T2D. We believe PHARMAC should strongly consider adding ethnicity criteria in all medications accessed through special authority.