Abstract
Aims:
Type 2 diabetes (T2D) and its complications disproportionally affect American Indian and Alaska Native (AI/AN) peoples. Prescribing decisions for anti-diabetic medications are complicated and require balancing medication benefits, costs and side effects. Little is known about trends in anti-diabetic medication use as well as acute diabetes complications among AI/AN adults. Here, we examined patterns and trends in anti-diabetic medication use and rates of hospital admissions or emergency department (ED) visits due to severe hypoglycaemia and hyperglycaemia among AI/AN adults with T2D.
Materials and Methods:
We conducted a retrospective analysis of Indian Health Service (IHS) Improving Health Care Delivery Data Project. A total of 39 183 AI/AN adults aged ≥18 years with T2D who used IHS or Tribal health services during any of the fiscal years (FYs) 2009–2013 were included. Utilization rates of each class of anti-diabetic medications and rates of severe hypoglycaemia and severe hyperglycaemia in emergency room and/or inpatient discharge diagnoses were calculated for each year. Longitudinal statistical models were fitted to examine time trends of anti-diabetic medication use and complications.
Results:
During 2009–2013, use of metformin (56.0%–60.5%), insulin (31.4%–35.9%) and dipeptidyl peptidase-4 inhibitors (1.4%–9.0%) increased, whereas the use of sulfonylureas (40.3%–32.9%) and thiazolidinediones (TZDs, 31.6%–8.8%) decreased significantly. Trends in severe hypoglycaemia (1.6%–0.8%) and severe hyperglycaemia (2.0%–1.6%) declined gradually.
Conclusions:
There were significant changes in the utilization of different anti-diabetic medication classes during 2009–2013 among AI/AN adults with T2D. Concurrently, there were significant reductions in severe hypoglycaemia and severe hyperglycaemia.
Keywords: anti-diabetic medications, severe hyperglycaemia, severe hypoglycaemia, T2D, trends
1 |. INTRODUCTION
Diabetes is a chronic health condition affecting more than 38 million people in the United States (US) in 2021 and is significantly associated with acute complications and higher mortality.1 Indian Health Service (IHS) and Tribal health programmes have implemented diabetes prevention and treatment programmes for American Indian and Alaska Native (AI/AN) peoples. AI/AN adults had the highest rates of diabetes in the nation: prevalence was 13.6% in 2019–2021, nearly twice that of non-Hispanic White adults (6.9%).2 Serious diabetes complications, such as diabetes-related end-stage renal disease, are also more common among AI/AN adults than non-Hispanic White adults.3
Treatment options for type 2 diabetes (T2D) have expanded considerably over the past few decades. In the 1990s, available medications included metformin, sulfonylureas, thiazolidinediones (TZDs) and insulin, which lower the haemoglobin A1c (HbA1c) by 0.5%–1.5%.4 However, sulfonylureas and insulin were found to be associated with weight gain and a higher risk of hypoglycaemia, the most common diabetes complication related to many poor outcomes, including falls, seizures or death.4–6 TZDs were reported to increase the risk of weight gain and cardiovascular diseases (CVDs).4,7 Since 2000, newer anti-diabetic medications have been approved for treating T2D, including dipeptidyl peptidase-4 inhibitor (DPP-4i), newer rapid- and long-acting insulin derivatives and several fixed-dose combinations.8 These newer medications have a much lower risk of causing hypoglycaemia and other adverse events, but their extremely high cost poses a challenge for patients and healthcare budgets.9
To improve care for individuals with T2D, the American Diabetes Association (ADA) and the European Association for the Study of Diabetes recommended metformin as the initial therapy for treating T2D in 2007, with insulin, sulfonylureas and TZDs as second-line therapies.10 In 2012, the ADA updated the guidelines to emphasize considering efficacy, costs and risks of hypoglycaemia and weight gain for add-on therapies.11 In light of the changes in T2D management strategies, it is crucial to examine the patterns and trends in the utilization of anti-diabetic medications. While several recent studies using US nationally representative data have reported trends in anti-diabetic medication use during 1999–2018,12–15 only one study has examined changes in medications and glycaemia among the American Indian (AI) population from 1975 to 2004,16 indicating a lack of knowledge about these trends among AI/AN adults.
Significant advances in diabetes management in recent decades have substantially reduced macrovascular and microvascular complications of diabetes. US national surveys and hospital discharge data have shown a decline in many diabetes-related complications from 1990 to 2010.17 Furthermore, among privately insured and US Medicare beneficiaries, rates of severe hypoglycaemia exceeded those of severe hyperglycaemia during 2001–2020, with a modest decline in hypoglycaemia hospitalizations.18,19 However, several recent studies found a resurgence in some diabetes-related complications during 2010–2020,19–21 including severe hyperglycaemia. Although previous studies have examined trends in diabetes-related complications, the rates and changes among AI/AN adults remain unknown.
In response to the heavy burden of diabetes in AI/AN peoples, IHS and Tribes developed approaches to improve diabetes care. In 1997, Congress created the Special Diabetes Program for Indians (SDPI), which provides funding for diabetes prevention and treatment to approximately 300 AI/AN programmes in 35 states as of 2014.22 In this study, we examined the patterns and trends in anti-diabetic medication use and rates of severe hypoglycaemia and severe hyperglycaemia which led to hospital admissions and emergency department (ED) visits from 2009 to 2013 among AI/AN adults aged 18 years or older with T2D using a national sample of IHS users.
2 |. MATERIALS AND METHODS
2.1 |. Data source
Approximately 27% (2.6 million) of individuals who self-identify as AI/AN in the US Census receive healthcare funded by IHS and Tribal health programmes, which include hospitals, clinics and health programmes operated by IHS (I), Tribal organizations (T) and urban Indian health programmes (U).23 Our study extracted data from the IHS Improving Health Care Delivery Data Project (IHS Data Project) which houses health status, utilization and treatment costs for over 640 000 AI/AN peoples, representing nearly 30% of AI/AN peoples who use IHS and Tribal services.24 The IHS Data Project includes a sample of data, collected from 15 IHS units (called project sites hereafter) located throughout the nation. These 15 sites had limited urban Indian clinic use during the study period; for this reason, we refer to them as IHS and Tribal (I/T, instead of I/T/U) providers in this study. This data infrastructure synthesizes existing electronic health record (EHR) data from numerous IHS platforms from October 1, 2007 to September 30, 2013 (fiscal year [FY] 2007–2013). Registration, demographic and I/T service data were extracted from the National Data Warehouse (NDW), while non-I/T provided but I/T paid service data were drawn from Purchased and Referred Care (PRC) services. More detail information on this data infrastructure is reported elsewhere.24
Project personnel partnered with IHS and Tribal organizations taking part in the IHS Data Project through a collaborative network which convenes regular meetings of three advisory committees (i.e., Steering, Project Site and Patient), travels to project sites and entails a process to obtain approvals from the IHS National Institutional Review Board (IRB); Tribal IRBs, Councils and Tribal Authorities and the University’s IRB (Colorado Multiple IRBs).
2.2 |. Study population
The IHS Data Project population is comparable to the national IHS population in age and sex. The study population included AI/AN adults (≥18 years old) with T2D who used IHS during any FY from 2009 to 2013. T2D patients were identified using validated algorithms based on diagnostic codes and medication codes.25 Data from six of the 15 project sites were excluded because two sites had low percentages of completed medication data, two did not provide I/T hospital inpatient and emergency services, one had incomplete PRC hospital data and one was not part of the IHS Data Project until FY2011. A total of 39 183 adults with T2D were included (Figure S1).
2.3 |. Main outcomes
We created dichotomous indicator variables for different classes of anti-diabetic medications, assigning a value of 1 if patients used the medications in each FY. We defined each class of anti-diabetic medication utilization rate in a FY as the proportion of patients who used each class of anti-diabetic medications in that year. Severe hypoglycaemia and severe hyperglycaemia events were defined as hospital admissions or ED visits with a principal discharge diagnosis code indicating hyperglycaemia18 or hypoglycaemia18,26 (Table S1). For each of these acute diabetes complications, we created a dichotomous indicator that equals 1 if a patient had at least one hospitalization or ED use due to that complication in a FY. We defined the observed rate of hospitalizations or ED visits due to hypoglycaemia and hyperglycaemia in a FY as the proportion of patients who had at least one complication event in that year.
2.4 |. Covariates
NDW data provided information on age, sex, body mass index (BMI) and health coverage. We employed ICD-9 diagnostic codes, supplemented by medication data, to create chronic condition variables that indicate an adult had or did not have a chronic disease. Sightlines™ DxCG Risk Solutions software groups ICD-9 codes into Diagnostic Cost Groups that are employed by the federal government and private insurers to identify chronic conditions.27 We used this software to identify adults with one or more types of comorbidities with T2D (Verisk Health, Inc. SightlinesTM DxCG Risk Solutions): CVDs, chronic kidney disease, depression, dementia, stroke, cognitive disorder, cancer, liver disease and hypertension as the serious chronic conditions. Using BMI, adults were classified as overweight, obese, normal and underweight.
2.5 |. Statistical analysis
Sample characteristics were described by the frequency and percentage distribution of age categories, sex, regions, BMI categories, healthcare coverage and numbers of chronic conditions. Bivariate analysis of the proportions of anti-diabetic medication use and rates of severe hypoglycaemia and hyperglycaemia by age group, sex and health coverage was also performed.
To examine time trends in the utilization rates of anti-diabetic medications and acute diabetes complications, generalized estimating equation models with a binomial distribution and logit link function were fitted, with an AR (1) covariance structure to account for repeated measures for the same individual. All models were adjusted for age, sex, health coverage (time-varying), chronic condition category (time-varying) and sites. Using FY2009 as the reference year, we included four dummy variables representing FY2010 to FY2013. Adjusted odds ratios (aORs) for each FY are reported. A two-sided p value < 0.05 was considered statistically significant. All statistical analyses were performed on SAS 9.4 (SAS, Inc. Cary, NC).
3 |. RESULTS
3.1 |. Sample characteristics
A total of 39 183 AI/AN adults with T2D were identified during FY2009–2013. The average age of the adults was 54.8 years, with 24%–26% in the 18–44 age group, 50%–51% in the 45–64 age group, 15%–16% in the 65–74 age group and 7%–8% in ≥75 age group. A little over half were female (~56% each FY). The proportion of patients covered by private insurance or Medicaid decreased gradually. Approximately two-thirds of the adults were classified as obese, 22%–24% overweight and 8%–9% normal or underweight. About 45% of the included patients had ≥2 concurrent chronic conditions, while ≤13% had no comorbidity (Table 1). The mean HbA1c values remained relatively stable over the study period (Figure S2).
TABLE 1.
Characteristics of AI/AN adults with T2D from FY2009 to FY2013.
| Characteristic—(N, %)/FY | FY2009 (N = 23 272) | FY2010 (N = 25 253) | FY2011 (N = 27 191) | FY2012 (N = 28 131) | FY2013 (N = 28 615) |
|---|---|---|---|---|---|
| Age category | |||||
| 18–44 | 6077 (26.11) | 6320 (25.03) | 6768 (24.89) | 6941 (24.67) | 6774 (23.67) |
| 45–64 | 11 742 (50.46) | 12 890 (51.04) | 13 938 (51.26) | 14 391 (51.16) | 14 621 (51.10) |
| 65–74 | 3702 (15.91) | 4068 (16.11) | 4292 (15.78) | 4521 (16.07) | 4803 (16.78) |
| > =75 | 1751 (7.52) | 1975 (7.82) | 2193 (8.07) | 2278 (8.10) | 2417 (8.45) |
| Sex | |||||
| Female | 13 186 (56.66) | 14 194 (56.21) | 15 268 (56.15) | 15 804 (56.18) | 16 063 (56.13) |
| Male | 10 086 (43.34) | 11 059 (43.79) | 11 923 (43.85) | 12 327 (43.82) | 12 552 (43.87) |
| Private health insurance (Yes) | 6149 (26.42) | 5849 (23.16) | 5603 (20.61) | 5458 (19.40) | 4033 (14.09) |
| Medicaid (Yes) | 5282 (22.70) | 5894 (23.34) | 5987 (22.02) | 5012 (17.82) | 4039 (14.11) |
| BMI categoriesa | N = 16 540 | N = 18 357 | N = 21 274 | N = 21 342 | N = 21 552 |
| Underweight (<18.5) | 78 (0.47) | 92 (0.50) | 99 (0.47) | 96 (0.45) | 92 (0.43) |
| Normal (18.5–25) | 1315 (7.95) | 1454 (7.92) | 1723 (8.10) | 1801 (8.44) | 1860 (8.63) |
| Overweight (25–30) | 3696 (22.35) | 4185 (22.80) | 4776 (22.45) | 5019 (23.52) | 5082 (23.58) |
| Obese (> = 30) | 11 451 (69.23) | 12 626 (68.78) | 14 676 (68.99) | 14 426 (67.59) | 14 518 (67.36) |
| Missing data of BMI categories | 6732 (28.93) | 6896 (27.31) | 5917 (21.76) | 6789 (24.13) | 7063 (24.68) |
| Regionsb | |||||
| East | 1565 (6.72) | 1642 (6.50) | 1721 (6.33) | 1746 (6.21) | 1762 (6.16) |
| Northern plains | 3248 (13.96) | 3414 (13.52) | 3764 (13.84) | 3828 (13.61) | 3819 (13.35) |
| Southern plains | 6750 (29.00) | 8101 (32.08) | 8938 (32.87) | 9090 (32.31) | 9111 (31.84) |
| Southwest | 11 709 (50.31) | 12 096 (47.90) | 12 768 (46.96) | 13 467 (47.87) | 13 923 (48.66) |
| Numbers of comorbidities | |||||
| 0 | 2732 (11.75) | 2948 (11.68) | 3188 (11.73) | 3503 (12.45) | 3690 (12.90) |
| 1 | 9683 (41.63) | 10 573 (41.88) | 11 407 (41.95) | 11 455 (40.72) | 11 636 (40.67) |
| 2 | 5115 (21.99) | 5732 (22.70) | 5922 (21.78) | 6095 (21.67) | 6104 (21.33) |
| > =3 | 5730 (24.63) | 5994 (23.74) | 6672 (24.54) | 7076 (25.16) | 7183 (25.10) |
| Anti-diabetic medicationsa | N = 22 903 | N = 24 696 | N = 26 330 | N = 27 116 | N = 26 020 |
| Any anti-diabetic med use | 19 843 (86.64) | 21 601 (87.47) | 23 007 (87.38) | 23 517 (86.73) | 22 304 (85.72) |
| Metformin use | 12 835 (56.04) | 14 754 (59.74) | 16 114 (61.20) | 16 446 (60.65) | 15 729 (60.45) |
| Insulin use | 7194 (31.41) | 7973 (32.28) | 8673 (32.94) | 9235 (34.06) | 9347 (35.92) |
| Sulfonylureas use | 9230 (40.30) | 9656 (39.10) | 9900 (37.60) | 9576 (35.31) | 8560 (32.90) |
| TZD use | 7244 (31.63) | 6885 (27.88) | 5801 (22.03) | 3749 (13.83) | 2284 (8.78) |
| DPP-4i use | 321 (1.40) | 532 (2.15) | 1515 (5.75) | 2357 (8.69) | 2333 (8.97) |
The total number of patients who had medication (or body mass index [BMI]) information were different from the total number of patients with T2D, because some patients’ medication (or BMI) information was missing.
Data from 6 of the 15 project sites were excluded because two sites had low percentages of completed medication data, two sites did not provide I/T hospital inpatient and emergency services, one site had incomplete purchased/referred care (PRC) hospital data and one was not part of IHS Data Project until FY2011.
3.2 |. Trends in anti-diabetic medication use: Overall versus by class
From FY2009 to FY2013, any anti-diabetic medication use decreased from 86.6% to 85.7% among AI/AN adults with T2D. Specifically, the use of metformin (56.0%–60.5%), insulin (31.4%–35.9%) and DDP-4i (1.4%–9.0%) increased, while the use of sulfonylurea (40.3%–32.9%) and TZDs (31.6%–8.8%) declined. Fixed-dose metformin combination use (0.02%–2.49%) increased, while fixed-dose metformin/sulfonylureas combination (4.1%–2.4%) declined (Figure 1A,B). The use of long-acting insulin (54.4%–68.5%) and rapid-acting insulin (23.0%–38.8%) increased, while short-acting insulin (27.0%–19.4%), intermediate-acting insulin (24.7%–12.9%) and premixed insulin (26.4%–19.1%) declined gradually (Figure 1C).
FIGURE 1.
Trends in utilization rates of anti-diabetic medications in American Indian and Alaska Native (AI/AN) adults with T2D from FY2009 to FY2013. (A) Trends in traditional anti-diabetic medication use; (B) trends in relatively new anti-diabetic medication use; (C) trends in insulin subtypes. 1Any class of anti-diabetic medication; 2any class of oral anti-diabetic medication; 3alogliptin-metformin HCl; linagliptin-metformin HCl; saxagliptin-metformin HCl; sitagliptin-metformin HCl; repaglinide-metformin HCl; canagliflozin-metformin HCl; 4glipizide-metformin HCl and glyburide-metformin. DDP-4i, dipeptidyl peptidase-4 inhibitors; TZDs, thiazolidinediones.
The youngest group (18–44) had the highest rate of using metformin among all age groups; 45–64 years old group had the highest rate of using insulin, while 45–74 years old group had the highest sulfonylureas or DPP-4i use. The overall patterns of time trends for males and females using metformin, insulin or DPP-4i were similar, while males had a higher rate of using sulfonylureas than females. The rates of metformin, sulfonylureas or DPP-4i use among Medicaid-enrolled patients were lower than that among those not enrolled in Medicaid, while the rate of insulin use was higher among Medicaid-enrolled patients compared with non-Medicaid enrolees. Patients with private insurance had a higher rate of metformin or DPP-4i use compared with those without private insurance, while the rate of insulin use was higher among those without private insurance (Figure S3). Southern Plains region had the highest rate of anti-diabetic medication use, followed by the Northern Plains region, the Southwest and the East region (Figure S4A).
Table 2 shows the adjusted regression models which indicate increased ORs for metformin, insulin and DPP-4i use when compared with FY2009. The highest aOR for metformin was observed in FY2011 (aOR = 1.20, 95% confidence interval (CI): 1.16–1.23), indicating a 20% increase in the odds of using metformin from FY2009 to FY2011. Additionally, private insurance and fewer comorbidities were significantly associated with higher odds of using metformin. The highest aOR for insulin use was in FY2013 (aOR = 1.43, 95% CI: 1.39–1.48). Medicaid enrolment and more comorbidities were significantly associated with higher odds of using insulin. Meanwhile, we found a decrease in the odds of sulfonylureas use from FY2009–2013, with the lowest aOR observed in FY2013 (aOR = 0.77, 95% CI: 0.74–0.79). Individuals with CVD, liver disease, dementia, alcohol/drug use disorders or chronic renal disease (CKD) had significantly lower odds of using metformin but higher odds of using insulin. AI/AN individuals with CVD or CKD had higher odds of DPP-4i use, while those with CVD had higher odds of sulfonylurea use. The highest aOR for DPP-4i use was observed in FY2013 (aOR = 8.68, 95% CI: 7.72–9.76). Patients with private health insurance were more likely to use DPP-4i.
TABLE 2.
Odds ratios (ORs) from unadjusted and adjusted generalized estimating equation models for anti-diabetic medications.
| Outcome | Metformin ever users |
Insulin ever users |
Sulfonylureas ever users |
DPP-4i ever users |
||||
|---|---|---|---|---|---|---|---|---|
|
|
Unadjusted |
Adjusted |
Unadjusted |
Adjusted |
Unadjusted |
Adjusted |
Unadjusted |
Adjusted |
| Covariates | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) |
| Year | ||||||||
| 2009 | Referent | Referent | Referent | Referent | Referent | Referent | Referent | Referent |
| 2010 | 1.11 (1.09, 1.13)*** | 1.13 (1.11, 1.16)*** | 1.09 (1.07, 1.11)*** | 1.08 (1.06, 1.11)*** | 0.95 (0.93, 0.96)*** | 0.94 (0.92, 0.96)*** | 1.42 (1.29, 1.56)*** | 1.42 (1.29, 1.56)*** |
| 2011 | 1.14 (1.11, 1.17)*** | 1.20 (1.16, 1.23)*** | 1.19 (1.16, 1.22)*** | 1.18 (1.15, 1.21)*** | 0.91 (0.89, 0.93)*** | 0.90 (0.88, 0.93)*** | 4.08 (3.65, 4.55)*** | 4.18 (3.74, 4.68)*** |
| 2012 | 1.09 (1.06, 1.12)*** | 1.17 (1.14, 1.20)*** | 1.31 (1.28, 1.35)*** | 1.29 (1.26, 1.33)*** | 0.84 (0.82, 0.86)*** | 0.83 (0.80, 0.85)*** | 6.63 (5.93, 7.42)*** | 7.13 (6.36, 7.99)*** |
| 2013 | 1.05 (1.02, 1.09)*** | 1.17 (1.13, 1.21)*** | 1.46 (1.42, 1.50)*** | 1.43 (1.39, 1.48)*** | 0.78 (0.75, 0.80)*** | 0.77 (0.74, 0.79)*** | 7.62 (6.80, 8.55)*** | 8.68 (7.72, 9.76)*** |
| Age (in years) × 10 years | 0.81 (0.80, 0.82)*** | 0.82 (0.81, 0.84)*** | 1.06 (1.04, 1.07)*** | 1.00 (0.99, 1.02) | 1.04 (1.03, 1.05)*** | 1.06 (1.05, 1.08)*** | 1.15 (1.13, 1.18)*** | 0.99 (0.96, 1.01) |
| Male | 1.00 (0.97, 1.04) | 0.99 (0.95, 1.03) | 0.95 (0.91, 0.99)** | 0.95 (0.92, 0.99)* | 1.24 (1.20, 1.29)*** | 1.24 (1.19, 1.28)*** | 1.05 (0.98, 1.12) | 0.93 (0.87, 1.08) |
| Private insurance (Yes) | 1.10 (1.07, 1.13)*** | 1.10 (1.07, 1.13)*** | 0.95 (0.93, 0.97)*** | 0.99 (0.96, 1.01) | 1.03 (1.00, 1.06)* | 1.01 (0.98, 1.04) | 0.92 (0.96, 0.98)** | 1.09 (1.03, 1.17)** |
| Medicaid (Yes) | 0.92 (0.90, 0.95)*** | 0.91 (0.88, 0.94)*** | 1.06 (1.03, 1.09)*** | 1.08 (1.05, 1.11)*** | 0.98 (0.95, 1.01) | 0.97 (0.94, 1.00) | 0.60 (0.56, 0.64)*** | 0.97 (0.88, 1.07) |
| Comorbidities | ||||||||
| Cardiovascular disease | 0.80 (0.78, 0.82)*** | 0.92 (0.90, 0.95)*** | 1.27 (1.24, 1.30)*** | 1.22 (1.19, 1.25)*** | 1.04 (1.02, 1.07)*** | 1.04 (1.01, 1.06)** | 1.17 (1.11, 1.23)*** | 1.12 (1.05, 1.19)*** |
| Liver disease | 0.81 (0.78, 0.85)*** | 0.82 (0.79, 0.86)*** | 1.21 (1.16, 1.26)*** | 1.16 (1.11, 1.21)*** | 1.02 (0.97, 1.06) | 1.03 (0.99, 1.08) | 1.06 (0.96, 1.17) | 1.03 (0.92, 1.16) |
| Depression | 1.00 (0.97, 1.03) | 1.01 (0.98, 1.05) | 1.17 (1.14, 1.21)*** | 1.16 (1.12, 1.19)*** | 0.98 (0.95, 1.01) | 1.00 (0.97, 1.03) | 1.14 (1.06, 1.22)*** | 0.98 (0.91, 1.05) |
| Dementia | 0.49 (0.44, 0.54)*** | 0.63 (0.57, 0.70)*** | 1.20 (1.07, 1.35)** | 1.01 (0.90, 1.14) | 0.68 (0.60, 0.77)*** | 0.67 (0.59, 0.75)*** | 0.82 (0.65, 1.03) | 0.75 (0.58, 0.98)* |
| Alcohol/drug use disorder | 0.89 (0.86, 0.93)*** | 0.87 (0.83, 0.90)*** | 1.10 (1.05, 1.14)*** | 1.05 (1.01, 1.09)* | 1.00 (0.96, 1.04) | 1.01 (0.97, 1.05) | 0.68 (0.62, 0.75)*** | 0.84 (0.75, 0.94)** |
| Chronic renal disease | 0.44 (0.42, 0.45)*** | 0.49 (0.47, 0.51)*** | 1.50 (1.45, 1.56)*** | 1.38 (1.33, 1.43)*** | 0.97 (0.93, 1.00) | 0.96 (0.92, 0.99)* | 1.27 (1.17, 1.37)*** | 1.15 (1.05, 1.25)** |
Abbreviation: DPP-4i, dipeptidyl peptidase-4 inhibitors.
p <0.05
p < 0.01
p < 0.001.
3.3 |. Trends in severe hypoglycaemia and severe hyperglycaemia
During FY2009-FY2013, the proportion of patients with severe hypoglycaemia declined gradually from 1.6% to 0.8% and the proportion of patients with severe hyperglycaemia decreased from 2.0% to 1.6% (Figure 2). The oldest group (≥75) had the highest rate of severe hypoglycaemia, while the youngest had the lowest rate of severe hypoglycaemia across all 5 years. Conversely, the youngest group had the highest rates of severe hyperglycaemia. The overall patterns of time trends of severe hypoglycaemia for males and females were similar, but males had a higher rate of hyperglycaemia than females. Rates of severe hypoglycaemia and hyperglycaemia among patients enrolled in Medicaid were around twice as high as those without Medicaid. Conversely, the rates of these complications were lower among those with private insurance than those without private insurance (Figure S5). East had the highest rates of severe hyperglycaemia and hypoglycaemia, followed by Northern Plains, Southwest and Southern Plains (Figure S4B,C).
FIGURE 2.
Observed rates of hospitalizations and emergency department (ED) visits due to severe hypoglycaemia and severe hyperglycaemia from FY2009 to FY2013.
The aORs for severe hyperglycaemia and severe hypoglycaemia were less than 1 in all the follow-up FYs compared with FY2009. The lowest aORs for severe hyperglycaemia and hypoglycaemia were observed in FY2011 (aOR = 0.78, 95% CI: 0.68–0.89) and in FY2013 (aOR = 0.49, 95% CI, 0.41, 0.58), respectively. Private insurance enrolment was significantly associated with lower odds of having severe hyperglycaemia and hypoglycaemia, while Medicaid enrolment and those with CVD, liver disease, dementia, depression, alcohol/drug use disorders or CKD were linked to higher odds of these complications (Table 3).
TABLE 3.
Odds ratios (ORs) from unadjusted and adjusted generalized estimating equation models for severe hyperglycaemia and severe hypoglycaemia.
| Outcome | Severe hyperglycaemia |
Severe hypoglycaemia |
||
|---|---|---|---|---|
|
|
Unadjusted |
Adjusted |
Unadjusted |
Adjusted |
| Covariates | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) |
| Year | ||||
| 2009 | Referent | Referent | Referent | Referent |
| 2010 | 0.94 (0.84, 1.07) | 0.96 (0.84, 1.08) | 0.83 (0.72, 0.96)* | 0.85 (0.73, 0.98)* |
| 2011 | 0.78 (0.68, 0.89)*** | 0.78 (0.68, 0.89)*** | 0.65 (0.56, 0.76)*** | 0.66 (0.57, 0.77)*** |
| 2012 | 0.80 (0.70, 0.91)*** | 0.81 (0.71, 0.93)** | 0.60 (0.52, 0.70)*** | 0.61 (0.52, 0.71)*** |
| 2013 | 0.80 (0.70, 0.91)*** | 0.83 (0.73, 0.95)** | 0.49 (0.41, 0.58)*** | 0.49 (0.41, 0.58)*** |
| Age (in years) × 10 years | 0.66 (0.64, 0.69)*** | 0.66 (0.63, 0.68)*** | 1.28 (1.23, 1.34)*** | 1.13 (1.07, 1.18)*** |
| Male | 1.35 (1.23, 1.49)*** | 1.19 (1.08, 131)*** | 0.91 (0.80, 1.02) | 0.84 (0.74, 0.95)** |
| Private insurance (Yes) | 0.73 (0.65, 0.82)*** | 0.81 (0.72, 0.91)*** | 0.72 (0.62, 0.84)*** | 0.86 (0.74, 1.01) |
| Medicaid (Yes) | 1.72 (1.55, 1.90)*** | 1.27 (1.13, 1.42)*** | 1.95 (1.73, 2.20)*** | 1.38 (1.21, 1.57)*** |
| Comorbidities | ||||
| Cardiovascular disease | 1.48 (1.34, 1.62)*** | 1.88 (1.68, 2.10)*** | 3.61 (3.22, 4.05)*** | 2.32 (2.02, 2.66)*** |
| Liver disease | 2.24 (1.96, 2.55)*** | 1.57 (1.37, 1.80)*** | 1.78 (1.48, 2.15)*** | 1.38 (1.14, 1.67)** |
| Depression | 1.37 (1.23, 1.52)*** | 1.19 (1.07, 1.33)** | 1.50 (1.32, 1.71)*** | 1.35 (1.18, 1.54)*** |
| Dementia | 1.10 (0.79, 1.53) | 1.45 (1.03, 2.04)* | 4.24 (3.31, 5.44)*** | 1.78 (1.37, 2.32)*** |
| Alcohol/drug use disorder | 3.45 (3.10, 3.84)*** | 2.02 (1.79, 2.28)*** | 2.13 (1.84, 2.46)*** | 2.01 (1.72, 2.36)*** |
| Chronic renal disease | 1.19 (1.05, 1.34)** | 1.27 (1.10, 1.45)*** | 3.65 (3.24, 4.11)*** | 2.11 (1.83, 2.43)*** |
p <0.05
p < 0.01
p < 0.001.
4 |. DISCUSSION
In a large and geographically diverse sample of AI/AN adults with T2D who used I/T services, we found the proportions of patients with metformin, insulin, DPP-4i and fixed-dose metformin combination use increased, while the use of sulfonylureas, TZDs and fixed-dose metformin/sulfonylureas declined from FY2009 to FY2013. Meanwhile, observed rates of patients with severe hypoglycaemia and severe hyperglycaemia declined gradually over 5 years.
Metformin remained the primary medication for T2D, consistent with the ADA guidelines.7,10,28 One study focusing on AI population found an increased use of any anti-diabetic medication and metformin,16 aligning with our findings. However, they observed an increase in TZDs use and no changes in sulfonylureas use from 1975 to 2004, which is consistent with trends before 2007 from studies using US national data.12,13 This is likely due to the absence of specific guidelines for anti-diabetic medications until 2007. Additionally, many studies have examined trends in anti-diabetic medication use among US general population during 2006–2018, and our findings are generally consistent with theirs. For instance, studies using privately insured and Medicare data15 or national health survey data12–14 reported similar trends: the use of metformin, insulin and DPP-4i increased, while that of sulfonylureas and TZDs decreased.
The reduced use of sulfonylureas may be due to their side effects, including hypoglycaemia and weight gain.7,28 TZD use also declined sharply because of concerns regarding increased risk of heart failure and weight gain.4,7 Furthermore, TZDs were found to be associated with an increased risk of bladder cancer and fractures.29,30 Conversely, DPP-4i is well tolerated and has a relatively lower risk of causing adverse events7,28 and thus gained popularity over the study period. Although insulin overall was associated with a potential risk of causing hypoglycaemia and weight gain,7,28 the subtypes of insulin used in basal-bolus insulin therapy (i.e., long-acting and rapid-acting insulin) were found to be the most effective, flexible and safe therapy compared with other insulin subtypes.31–34 In our study, although overall use of insulin increased, long-acting and rapid-acting insulins had the highest utilization rate and predominantly increased over the study period, suggesting more optimal glycaemic management with insulin in this population. Other factors such as health interventions of public health campaigns and changes in clinical guidelines could also contribute to the changes in trends and patterns of anti-diabetic medication use.
Males had a higher use of sulfonylureas, likely due to better glycaemia response compared with females.35 All patients had access to health services through IHS and Tribal health programmes, but the type of health coverage influenced medication use and risk of acute diabetes complications. Medicaid patients had a much lower DPP-4i use and higher rates of severe hypoglycaemia and hyperglycaemia. In contrast, private insurance was linked to a higher DPP-4i use and lower complication rates. The disparities suggest that income may be associated with the use of newer, but more expensive, anti-diabetic medications and the risk of these acute diabetes complications. This aligns with studies reporting that avoidable hospitalizations, including hospital admissions for uncontrolled diabetes and short-term complications, decreased as community income levels increased in various race/ethnic groups.36
Consistent with most prior studies,15,18,37 rates of severe hypoglycaemia were highest among the oldest age group across all 5 years. This might be partially caused by a greater prevalence of certain comorbidities that increase the risk of the events among older adults. In contrast to severe hypoglycaemia, rates of severe hyperglycaemia were highest among the youngest age group and decreased gradually with increased age; this is not unexpected since other studies showed that these complications are often associated with youth-onset diabetes.38 Although the overall rate of severe hyperglycaemia declined over time, it remained substantially higher in the youngest group and seemed to increase in the last year of data included. To keep reducing rates of this complication in young AI/AN individuals with diabetes, future diabetes management strategies could enhance education on medication adherence and sick day management among younger patients.39
Our study is among the first to report improvements in severe hypoglycaemia and hyperglycaemia among AI/AN adults with T2D. The decline in hypoglycaemia is likely linked to changes in anti-diabetic medication use, such as reduced sulfonylureas and increased DPP-4i use. DPP-4i has been shown to substantially lower risk of causing hypoglycaemia than sulfonylureas.40 Additionally, although insulin use increased over all 5 years, safer insulin types associated with fewer severe hypoglycaemia events and less weight gain increased predominately, contributing to the reduced severe hypoglycaemia rates. Other contributing factors include the shift in guidelines towards individualized glycaemic targets, increased emphasis on team-based case and heightened education and knowledge about interventions for these complications among both healthcare providers and patients.7,41
Analyses of national data revealed dramatic declining trends in hospital admissions for hyperglycaemia among US older adults with diabetes from 1999 to 2020, while hypoglycaemia hospitalizations only decreased slightly.18,19 During 2009–2015, hospital utilization for hyperglycaemia among the US general population with diagnosed diabetes resurged by >70%,21 while the rates of severe hypoglycaemia decreased by <10% from 2009 to 2013.15 The reasons for the recent increase in these diabetes-related complications were unclear but may be linked to rising diabetes prevalence42 and changes in socioeconomic factors after the 2008 economic recession.21 Investigating if there was a resurgence of hyperglycaemia and/or hypoglycaemia among AI/AN adults after 2013 represents a critical next step of the current study.
This study has some limitations. First, the characteristics of our sample might change over time. As health insurance status is correlated with different anti-diabetic medication use and diabetes acute complications, the trends in medication use and these complications in our study may be partially associated with changes in health insurance coverage over time. However, health insurance coverage was considered a time-varying covariate in our final regression models, which substantially mitigated this concern. Second, some IHS and Tribal health programmes have limited resources to add newer and expensive anti-diabetic medications to patients’ formularies. Adults with private health coverage may have access to these medications outside of IHS sites. Third, some crucial variables or potential confounders, including education level and total income, lifestyle factors, medication adherence and persistence, provider practices or mortality related data, were not available in our data. Fourth, since we do not have access to IHS data after 2013, some newer anti-diabetic medications were unavailable during the study period. Our current findings need to be confirmed and enhanced by obtaining related data after FY2013 to investigate time trends across a longer term or period. Lastly, hypoglycaemic and hyperglycaemic events that did not lead to being hospitalized or receiving ED services could not be identified using the available data.
Despite the above limitations, our study is the first to examine the patterns and trends of anti-diabetic medication use and rates of severe hypoglycaemia and hyperglycaemia among AI/AN adults with T2D. Additionally, our findings are based on EHR data from a large, geographically diverse sample of AI/AN adults who use I/T services, which may be a good reference for other healthcare systems, especially those that serve populations with limited resources.
In conclusion, the patterns of anti-diabetic medication use changed substantially among AI/AN adults with T2D from FY2009 to FY2013. The utilization rates of anti-diabetic medications that can cause hypoglycaemia or other adverse events decreased dramatically, while those of safer anti-diabetic medications increased significantly. Additionally, trends in severe hyperglycaemia and hypoglycaemia declined during those 5 years, supporting the effectiveness of the systematic diabetes management approach implemented by SDPI in the past few decades. The multidisciplinary team-based and evidence-informed approaches adopted by IHS and its Tribal partners, such as providing health education, clinical pharmacy and care management services, have resulted in remarkable improvements in many diabetes care outcomes. However, rates of acute complications for IHS users with Medicaid enrolment were much higher than those of IHS users without Medicaid, indicating low income may be a barrier to better diabetes outcomes.
Supplementary Material
ACKNOWLEDGEMENTS
The data used in this secondary analysis stem from the Indian Health Service (IHS) Health Care Delivery Data Project. The data set includes information for many American Indian and Alaska Native communities. This work was conducted with the guidance and advice of IHS and Tribal health programme colleagues, as well as members of the project’s Steering, Project Site and Patient Committees. Members of Tribal and IHS National Institutional Review Board (IRB), Tribal IRBs, Councils and Tribal Authorities educate us about the health concerns they have for their Tribal members and how they hope this project will inform their work. This project relies on their support and approval.
FUNDING INFORMATION
This study was supported by the National Institute on Aging (R01AG061189), National Institute of Diabetes and Digestive and Kidney Diseases (P30DK092923) and National Institute on Minority Health and Health Disparities (U54MD000507). Funding for the development of the data infrastructure was supported by the Patient-Centered Outcomes Research Institute (AD-1304-6451) and Agency for Healthcare Research and Quality (290-2006-00020-I).
Patient-Centered Outcomes Research Institute, Grant/Award Number: AD-1304-6451; National Institute of Diabetes and Digestive and Kidney Diseases, Grant/Award Number: P30DK092923; National Institute on Minority Health and Health Disparities, Grant/Award Number: U54MD000507; Agency for Healthcare Research and Quality, Grant/Award Number: 290-2006-00020-I; National Institute on Aging, Grant/Award Number: R01AG061189
Footnotes
Disclaimer: The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the Indian Health Service or the funding agencies.
SUPPORTING INFORMATION
Additional supporting information can be found online in the Supporting Information section at the end of this article.
CONFLICT OF INTEREST STATEMENT
The authors claim no conflicts of interests.
DATA AVAILABILITY STATEMENT
The data from the IHS Data Project used to support the findings of this study have not been made available because of IHS and Tribal regulations regarding data confidentiality and security.
REFERENCES
- 1.National Diabetes Statistics Report. Centers for Disease Control and Prevention. 2024. https://www.cdc.gov/diabetes/php/data-research/index.html.
- 2.CDC, National Diabetes Statistics Report. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention. CDC, National Diabetes Statistics Report; 2021. [Google Scholar]
- 3.IHS, Diabetes in American Indians and Alaska Natives: Facts At-a-Glance. Indian Health Service. Department of Health and Human Services; 2012. [Google Scholar]
- 4.White JR Jr. A brief history of the development of diabetes medications. Diabetes Spectr. 2014;27(2):82–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Zoungas S, Patel A, Chalmers J, et al. Severe hypoglycemia and risks of vascular events and death. N Engl J Med. 2010;363(15):1410–1418. [DOI] [PubMed] [Google Scholar]
- 6.Bonds DE, Miller ME, Bergenstal RM, et al. The association between symptomatic, severe hypoglycaemia and mortality in type 2 diabetes: retrospective epidemiological analysis of the ACCORD study. BMJ. 2010;340:340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of hyperglycemia in type 2 diabetes: a patient-centered approach: position statement of the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Spect. 2012; 25(3):154–171. [DOI] [PubMed] [Google Scholar]
- 8.Dahlén AD, Dashi G, Maslov I, et al. Trends in anti-diabetic drug discovery: FDA approved drugs, new drugs in clinical trials and global sales. Front Pharmacol. 2022;12:4119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Committee ADAPP. 9. Pharmacologic approaches to glycemic treatment: standards of care in diabetes—2024. Diabetes Care. 2023;47-(Supplement_1):S158–S178. doi: 10.2337/dc24-S009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Association AD. Standards of medical care in diabetes—2007. Diabetes care. 2007;30(Suppl 1):S4–S41. [DOI] [PubMed] [Google Scholar]
- 11.Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of hyperglycemia in type 2 diabetes: a patient-centered approach: position statement of the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care. 2012; 35(6):1364–1379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Fang M, Wang D, Coresh J, Selvin E. Trends in diabetes treatment and control in US adults, 1999–2018. N EnglJ Med. 2021;384(23):2219–2228. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Le P, Chaitoff A, Misra-Hebert AD, Ye W, Herman WH, Rothberg MB. Use of antihyperglycemic medications in US adults: an analysis of the National Health and Nutrition Examination Survey. Diabetes Care. 2020;43(6):1227–1233. [DOI] [PubMed] [Google Scholar]
- 14.Raval AD, Vyas A. National trends in diabetes medication use in the United States: 2008 to 2015. J Pharm Pract. 2020;33(4):433–442. [DOI] [PubMed] [Google Scholar]
- 15.Lipska KJ, Yao X, Herrin J, et al. Trends in drug utilization, glycemic control, and rates of severe hypoglycemia, 2006–2013. Diabetes Care. 2017;40(4):468–475. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Looker HC, Krakoff J, Andre V, et al. Secular trends in treatment and control of type 2 diabetes in an American Indian population: a 30-year longitudinal study. Diabetes Care. 2010;33(11):2383–2389. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Gregg EW, Li Y, Wang J, et al. Changes in diabetes-related complications in the United States, 1990–2010. N Engl J Med. 2014;370(16):1514–1523. [DOI] [PubMed] [Google Scholar]
- 18.Lipska KJ, Ross JS, Wang Y, et al. National trends in US hospital admissions for hyperglycemia and hypoglycemia among Medicare beneficiaries, 1999 to 2011. JAMA Int Med. 2014;174(7):1116–1124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.McCoy RG, Herrin J, Galindo RJ, et al. Rates of hypoglycemic and hyperglycemic emergencies among US adults with diabetes, 2011–2020. Diabetes Care. 2023;46(2):e69–e71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Geiss LS, Li Y, Hora I, Albright A, Rolka D, Gregg EW. Resurgence of diabetes-related nontraumatic lower-extremity amputation in the young and middle-aged adult US population. Diabetes Care. 2019;42(1):50–54. [DOI] [PubMed] [Google Scholar]
- 21.Gregg EW, Hora I, Benoit SR. Resurgence in diabetes-related complications. Jama. 2019;321(19):1867–1868. [DOI] [PubMed] [Google Scholar]
- 22.Indian Health Service, Special Diabetes Program for Indians. 2020. Report to Congress. Changing the course of diabetes: Turning hope into reality. [PubMed] [Google Scholar]
- 23.US Census Bureau. American Indian and Alaska Native Heritage Month: November 2017. US Department of Commerce; 2017. Accessed November 2017 https://www.census.gov/newsroom/facts-for-features/2017/aian-month.html [Google Scholar]
- 24.O’Connell J, Guh S, Ouellet J, et al. ARRA ACTION: Comparative Effectiveness of Health Care Delivery Systems for American Indians and Alaska Natives Using Enhanced Data Infrastructure: Final Report. Rockville, MD: Agency for Healthcare Research and Quality. 2014. [Google Scholar]
- 25.Dai J, Niu X, Bullock A, Manson SM, O’Connell J, Jiang L. Prevalence of Diagnosed Type 1 and Type 2 Diabetes Among American Indian and Alaska Native Peoples in 2012–2013. Diabetes Care. 2024;47(1):e1–e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Geller AI, Shehab N, Lovegrove MC, et al. National estimates of insulin-related hypoglycemia and errors leading to emergency department visits and hospitalizations. JAMA Int Med. 2014;174(5):678–686. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Verisk Health, Inc. Sightlines™ DxCG Risk Solutions. Version 4.0.1. 2011. [Google Scholar]
- 28.Association AD. 9. Pharmacologic approaches to glycemic treatment: Standards of Medical Care in Diabetes—2020. Diabetes Care. 2020; 43(Supplement_1):S98–S110. [DOI] [PubMed] [Google Scholar]
- 29.Tuccori M, Filion KB, Yin H, Oriana HY, Platt RW, Azoulay L. Pioglitazone use and risk of bladder cancer: population based cohort study. BMJ. 2016;352. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Loke YK, Singh S, Furberg CD. Long-term use of thiazolidinediones and fractures in type 2 diabetes: a meta-analysis. CMAJ. 2009;180(1):32–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Candido R, Wyne K, Romoli E. A review of basal-bolus therapy using insulin glargine and insulin lispro in the management of diabetes mellitus. Diabetes Ther. 2018;9:927–949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Ampudia-Blasco FJ, Rossetti P, Ascaso JF. Basal plus basal–bolus approach in type 2 diabetes. Diabetes Technol Therap. 2011;13(S1):S-75–S-83. [DOI] [PubMed] [Google Scholar]
- 33.Torres Roldan VD, Urtecho M, Nayfeh T, et al. A systematic review supporting the Endocrine Society Guidelines: management of diabetes and high risk of hypoglycemia. J Clin Endocrinol Metab. 2023; 108(3):592–603. [DOI] [PubMed] [Google Scholar]
- 34.Bradley MC, Chillarige Y, Lee H, et al. Severe hypoglycemia risk with long-acting insulin analogs vs neutral protamine Hagedorn insulin. JAMA Int Med. 2021;181(5):598–607. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Dennis JM, Henley WE, Weedon MN, et al. Sex and BMI alter the benefits and risks of sulfonylureas and thiazolidinediones in type 2 diabetes: a framework for evaluating stratification using routine clinical and individual trial data. Diabetes Care. 2018;41(9):1844–1853. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Correa-de-Araujo R, McDermott K, Moy E. Gender differences across racial and ethnic groups in the quality of care for diabetes. Women’s Health Issues. 2006;16(2):56–65. [DOI] [PubMed] [Google Scholar]
- 37.Zhong VW, Juhaeri J, Cole SR, et al. Incidence and trends in hypoglycemia hospitalization in adults with type 1 and type 2 diabetes in England, 1998–2013: a retrospective cohort study. Diabetes Care. 2017;40(12):1651–1660. [DOI] [PubMed] [Google Scholar]
- 38.Desai D, Mehta D, Mathias P, Menon G, Schubart UK. Health care utilization and burden of diabetic ketoacidosis in the US over the past decade: a nationwide analysis. Diabetes Care. 2018;41(8):1631–1638. [DOI] [PubMed] [Google Scholar]
- 39.Kitabchi AE, Umpierrez GE, Miles JM, Fisher JN. Hyperglycemic crises in adult patients with diabetes. Diabetes Care. 2009;32(7): 1335–1343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Mishriky BM, Cummings DM, Tanenberg RJ. The efficacy and safety of DPP4 inhibitors compared to sulfonylureas as add-on therapy to metformin in patients with Type 2 diabetes: A systematic review and meta-analysis. Diabetes Res Clin Pract. 2015;109(2):378–388. doi: 10.1016/j.diabres.2015.05.025 [DOI] [PubMed] [Google Scholar]
- 41.ElSayed NA, Aleppo G, Aroda VR, et al. 9. Pharmacologic approaches to glycemic treatment: standards of care in diabetes—2023. Diabetes Care. 2023;46(Supplement_1):S140–S157. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Sharma M, Nazareth I, Petersen I. Trends in incidence, prevalence and prescribing in type 2 diabetes mellitus between 2000 and 2013 in primary care: a retrospective cohort study. BMJ Open. 2016;6(1):e010210. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data from the IHS Data Project used to support the findings of this study have not been made available because of IHS and Tribal regulations regarding data confidentiality and security.