Over the past two decades, many new treatment options have been introduced for type 2 diabetes. Increasingly, there is debate regarding the effectiveness of insulin as the last-line therapy. While diabetes control has improved over the past two decades (1–3), a substantial portion of patients is still not meeting the clinical targets. To examine national trends in glycemic control by insulin and oral diabetes medication use, we conducted a cross-sectional study of 4,947 adult participants with diagnosed diabetes (all types) in the National Health and Nutrition Examination Survey (NHANES) in the 1988–1994 and 1999–2012 cycles. The NHANES surveys are cross-sectional, complex probability samples of the U.S. civilian noninstitutionalized population. All NHANES protocols were approved by an institutional review board, and informed consent was obtained from participants. Hemoglobin A1c (HbA1c) was measured in whole blood samples using high-performance liquid chromatography methods performed on instruments certified by the NGSP and standardized to the reference method used in the Diabetes Control and Complications Trial (DCCT). We calibrated the HbA1c values to account for changes in measurement methods over the 20 years of data collection in NHANES (1). All statistical analyses accounted for the complex sampling design of the surveys and incorporated sampling weights to generate nationally representative estimates.
The proportion of patients with diabetes currently on any insulin (includes insulin only and insulin plus oral diabetes medications) was 29.1% (95% CI 26.7–31.5) in 2005–2012 and has been relatively stable since 1988 with 30.3% (95% CI 26.4–34.5) of patients in 1988–1994 and 24.8% (95% CI 21.2–28.8) in 1999–2004 (Table 1). Duration of diabetes was a strong, independent determinant of insulin use (data not shown), and over the past 24 years the duration of diabetes for those on any insulin has increased significantly (12.7 to 17.3 years, P < 0.001) compared with the duration of diabetes for those on oral medication only (7.0 to 8.5 years, P = 0.36). In 2005–2012, the prevalence of any insulin use among adults aged 65 years or older was almost identical to adults aged <65 years (27.8% vs. 29.8%, P = 0.391). Further, the current (2005–2015) prevalence of insulin-only use among adults 65 or older was 12.2% (95% CI 9.8–15.0) compared with 15.4% (95% CI 12.8–18.3) in adults <65 years (P = 0.20). Among whites in 2005–2012, the prevalence of any insulin use was 31.8% (95% CI 28.5–35.3) compared with 30.8% (95% CI 26.8–35.0) in African Americans and only 22% (95% CI 17.4–27.3) in Mexican Americans. After adjustment for duration of diabetes, age, sex, BMI, and education, Mexican Americans were still significantly less likely to receive insulin therapy as compared with whites (data not shown). In 2005–2012, among patients who had any insulin use, 31.4% (95% CI 27.0–36.2) had an HbA1c <7% and 61.7% (95% CI 56.5–66.6) had an HbA1c <8%.
Table 1.
Trends in insulin use and glycemic control by insulin use, according to age, race, and education, for U.S. adults with diagnosed diabetes, NHANES 1988–1994, 1999–2004, and 2005–2012
| 1988–1994 (N = 1,205) | 1999–2004 (N = 1,317) | 2005–2012 (N = 2,425) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Insulin only | No insulin (oral only) | Insulin and oral | No medication | Insulin only | No insulin (oral only) | Insulin and oral | No medication | Insulin only | No insulin (oral only) | Insulin and oral | No medication | |
| Overall | ||||||||||||
| % (unweighted n) | 26.8 (323) | 43.3 (540) | 3.5 (50) | 26.4 (292) | 15.0 (183) | 56.7 (799) | 9.8 (134) | 18.5 (201) | 14.1 (324) | 56.9 (1,437) | 14.9 (347) | 14.1 (317) |
| Age (years) | ||||||||||||
| 20 to <50 | 27.3 (6.4) | 39.4 (5.6) | 5.0 (2.5) | 28.3 (4.2) | 17.0 (3.2) | 52.2 (4.2) | 6.3 (2.1) | 24.5 (3.4) | 21.8 (2.4) | 47.3 (3.1) | 8.8 (1.5) | 22.0 (2.6) |
| 50 to <65 | 21.2 (2.8) | 47.3 (4.5) | 3.9 (1.2) | 27.6 (3.7) | 10.8 (1.8) | 58.5 (3.1) | 13.1 (2.2) | 17.6 (2.1) | 11.5 (1.6) | 58.6 (2.5) | 17.8 (1.6) | 12.1 (1.4) |
| ≥65 | 31.1 (2.5) | 42.2 (2.7) | 2.3 (0.8) | 24.4 (2.4) | 17.8 (2.2) | 57.8 (2.6) | 8.9 (1.3) | 15.5 (2.2) | 12.2 (1.3) | 60.8 (2.1) | 15.7 (1.5) | 11.4 (1.3) |
| Race | ||||||||||||
| Non-Hispanic white | 26.4 (2.9) | 43.0 (3.3) | 2.9 (0.8) | 27.7 (2.9) | 16.7 (2.4) | 55.7 (2.8) | 8.9 (1.6) | 18.8 (2.8) | 15.8 (1.5) | 55.1 (2.2) | 16.0 (1.2) | 13.1 (1.7) |
| Non-Hispanic black | 36.9 (3.1) | 35.5 (3.0) | 6.6 (1.6) | 21.0 (2.1) | 18.7 (2.2) | 54.0 (3.1) | 13.4 (2.3) | 13.9 (2.6) | 15.7 (1.7) | 56.1 (2.2) | 15.1 (1.4) | 13.1 (1.4) |
| Mexican American | 18.6 (2.2) | 52.2 (2.4) | 3.8 (0.9) | 25.4 (2.3) | 9.4 (2.3) | 66.5 (2.2) | 8.8 (2.1) | 15.3 (2.1) | 10.0 (1.9) | 60.6 (2.6) | 12.0 (2.0) | 17.4 (2.2) |
| Education | ||||||||||||
| High school or less | 27.4 (1.9) | 44.8 (3.0) | 3.4 (1.0) | 24.5 (2.4) | 16.4 (2.1) | 56.7 (2.2) | 9.8 (1.4) | 17.1 (1.8) | 12.7 (1.5) | 58.9 (2.0) | 14.2 (1.4) | 14.1 (1.5) |
| Some college | 28.3 (9.7) | 38.5 (7.5) | 4.7 (3.1) | 28.5 (7.4) | 12.7 (2.1) | 59.0 (3.8) | 9.9 (2.1) | 18.4 (3.3) | 16.3 (2.1) | 51.9 (3.2) | 16.6 (2.0) | 15.1 (2.1) |
| College or more | 21.4 (5.1) | 39.7 (8.1) | 1.2 (1.0) | 37.7 (8.0) | 13.9 (4.4) | 52.1 (5.2) | 9.6 (2.4) | 24.4 (4.4) | 14.8 (2.6) | 58.6 (3.8) | 14.5 (2.8) | 12.1 (2.3) |
| Duration of diabetes, mean (SE) | 13.0 (0.7) | 7.0 (0.5) | 10.0 (2.0) | 9.9 (0.7) | 18.6 (1.2) | 10.4 (0.6) | 15.6 (1.5) | 8.7 (1.0) | 19.3 (1.0) | 8.5 (0.3) | 15.4 (0.7) | 8.7 (0.8) |
| Glycemic control | ||||||||||||
| Calibrated HbA1c <7% | ||||||||||||
| % (unweighted n) | 34.7 (111) | 43.9 (243) | 25.2 (13) | 82.0 (236) | 29.8 (54) | 49.4 (383) | 28.2 (34) | 78.3 (153) | 34.0 (111) | 65.7 (907) | 29.0 (106) | 80.8 (241) |
| Age (years) | ||||||||||||
| 20 to <50 | 21.0 (6.5) | 35.4 (10.2) | 29.1 (24.9) | 85.3 (7.1) | 11.1 (4.6) | 35.4 (5.0) | 28.2 (11.9) | 73.4 (8.2) | 30.3 (6.9) | 56.2 (5.0) | 12.2 (5.1) | 79.0 (4.6) |
| 50 to <65 | 35.4 (7.4) | 40.9 (5.2) | 17.3 (9.1) | 78.3 (8.8) | 36.2 (8.3) | 47.8 (4.2) | 27.9 (7.4) | 75.1 (8.0) | 30.4 (6.5) | 63.5 (2.5) | 22.9 (5.3) | 72.8 (5.8) |
| ≥65 | 41.4 (5.3) | 51.4 (3.6) | 31.5 (15.4) | 83.4 (5.8) | 37.2 (7.0) | 58.6 (3.2) | 28.8 (8.7) | 86.4 (4.0) | 41.0 (5.2) | 72.1 (2.2) | 41.1 (5.0) | 90.9 (2.5) |
| Race | ||||||||||||
| Non-Hispanic white | 33.0 (4.9) | 45.2 (4.5) | 29.2 (15.4) | 81.4 (6.0) | 30.3 (6.6) | 56.7 (2.6) | 34.7 (6.6) | 84.4 (4.9) | 34.9 (4.6) | 69.9 (2.4) | 30.8 (4.7) | 89.9 (3.1) |
| Non-Hispanic black | 36.9 (6.1) | 37.5 (5.4) | 22.8 (11.0) | 83.4 (4.4) | 28.4 (6.8) | 41.8 (3.7) | 21.1 (6.6) | 63.9 (8.2) | 36.4 (5.6) | 62.7 (2.6) | 28.6 (4.7) | 69.0 (4.9) |
| Mexican American | 27.7 (7.2) | 38.4 (4.6) | 28.7 (12.3) | 69.9 (5.8) | 26.3 (10.4) | 30.9 (2.6) | 26.5 (8.2) | 66.2 (8.5) | 22.1 (6.4) | 49.8 (3.4) | 26.1 (8.0) | 60.7 (8.3) |
| Education | ||||||||||||
| High school or less | 37.1 (4.7) | 47.2 (4.2) | 15.4 (6.6) | 84.2 (4.4) | 28.3 (6.4) | 46.5 (2.4) | 25.3 (5.9) | 79.9 (5.2) | 28.5 (3.8) | 65.5 (2.4) | 34.7 (4.1) | 73.7 (3.7) |
| Some college | 20.8 (12.8) | 28.2 (8.0) | 55.3 (34.2) | 89.0 (6.3) | 34.5 (10.6) | 51.5 (4.4) | 27.3 (9.3) | 69.9 (8.5) | 35.2 (6.1) | 62.4 (4.0) | 24.1 (6.0) | 91.0 (3.6) |
| College or more | 41.4 (12.8) | 33.4 (11.1) | 11.9 (14.0) | 73.3 (17.5) | 28.8 (15.0) | 58.2 (5.3) | 42.5 (13.3) | 85.8 (8.1) | 45.8 (8.8) | 71.1 (4.0) | 20.9 (7.0) | 86.3 (4.4) |
| Duration of diabetes, mean (SE) | 12.9 (1.0) | 6.3 (0.7) | 9.3 (1.5) | 10.7 (0.9) | 18.9 (1.6) | 10.9 (0.9) | 14.6 (2.3) | 8.7 (1.1) | 19.8 (1.6) | 8.2 (0.4) | 16.1 (1.2) | 9.1 (1.0) |
| Calibrated HbA1c <8% | ||||||||||||
| % (unweighted n) | 50.9 (168) | 63.5 (325) | 63.1 (28) | 89.5 (259) | 55.9 (106) | 73.8 (578) | 50.8 (71) | 84.6 (170) | 62.4 (195) | 83.2 (1,168) | 60.9 (207) | 88.5 (273) |
| Age (years) | ||||||||||||
| 20 to <50 | 31.1 (9.0) | 50.9 (7.6) | 70.8 (22.1) | 87.7 (7.1) | 31.4 (8.6) | 58.2 (5.5) | 37.7 (13.5) | 80.1 (7.6) | 54.8 (6.3) | 70.6 (4.7) | 24.5 (8.6) | 85.1 (4.0) |
| 50 to <65 | 51.6 (7.0) | 61.6 (5.7) | 64.2 (14.0) | 82.7 (8.2) | 57.5 (8.6) | 71.7 (3.5) | 47.0 (9.2) | 81.8 (7.3) | 60.7 (7.0) | 81.0 (2.5) | 56.4 (5.1) | 83.0 (5.3) |
| ≥65 | 60.6 (5.3) | 72.3 (3.6) | 51.8 (20.2) | 97.2 (1.5) | 69.3 (5.5) | 84.3 (2.8) | 61.7 (8.3) | 91.7 (3.5) | 71.9 (5.1) | 90.9 (1.0) | 77.6 (5.1) | 97.8 (0.9) |
| Race | ||||||||||||
| Non-Hispanic white | 49.0 (6.1) | 67.6 (4.5) | 76.7 (15.1) | 89.1 (5.3) | 57.3 (5.0) | 78.2 (2.6) | 54.2 (72.4) | 90.0 (4.2) | 66.5 (4.5) | 87.5 (1.7) | 66.2 (4.7) | 94.2 (3.0) |
| Non-Hispanic black | 54.3 (6.3) | 48.9 (5.1) | 53.3 (15.1) | 91.4 (3.7) | 46.6 (7.2) | 68.1 (3.9) | 48.8 (7.9) | 83.1 (5.7) | 53.5 (5.4) | 78.9 (2.3) | 56.2 (5.5) | 81.7 (4.2) |
| Mexican American | 53.8 (8.7) | 57.6 (5.1) | 51.3 (11.6) | 81.4 (5.2) | 61.0 (7.7) | 55.3 (4.1) | 58.9 (9.1) | 73.4 (8.7) | 49.9 (6.9) | 71.3 (4.1) | 56.1 (8.5) | 71.5 (6.2) |
| Education | ||||||||||||
| High school or less | 52.5 (4.4) | 67.3 (4.1) | 56.8 (14.6) | 93.3 (3.0) | 52.5 (5.9) | 71.1 (2.3) | 46.9 (6.1) | 82.5 (5.0) | 60.3 (4.4) | 82.8 (1.9) | 68.3 (4.4) | 82.8 (3.5) |
| Some college | 32.9 (16.5) | 60.9 (10.1) | 55.3 (34.2) | 98.2 (1.3) | 47.3 (10.7) | 76.4 (3.5) | 50.2 (10.1) | 80.7 (7.7) | 60.9 (6.4) | 81.3 (3.1) | 50.2 (6.0) | 95.5 (2.2) |
| College or more | 51.2 (12.0) | 45.2 (15.3) | 88.7 (13.3) | 73.5 (17.5) | 86.9 (7.8) | 80.3 (4.9) | 68.7 (13.6) | 95.9 (4.6) | 70.1 (8.4) | 87.0 (2.8) | 58.3 (10.7) | 94.6 (2.6) |
| Duration of diabetes, mean (SE) | 13.2 (0.9) | 6.9 (0.6) | 9.4 (2.6) | 10.3 (0.8) | 20.1 (1.0) | 10.8 (0.7) | 16.8 (1.7) | 8.3 (1.0) | 20.4 (1.3) | 8.5 (0.4) | 16.5 (0.8) | 8.9 (0.9) |
Data are presented as weighted percent (SE), unless otherwise noted. Results for “Other” race/ethnicity are not included here because of small numbers. A total of 77 people were missing education level (for analyses stratified by education level, overall, N = 4,870; 1988–1994, n = 1,134; 1999–2004, n = 1,316; 2005–2012, n = 2,420).
Insulin is the last-line therapy for many patients. Our results suggest improvements are needed in patient and professional education on insulin use, new insulin formulations and delivery technology, and/or systems of care to improve glycemic control in patients receiving insulin.
Article Information
Funding. E.S. was supported by the National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health grant K24DK106414. C.M.P. was supported by the National Heart, Lung, and Blood Institute, National Institutes of Health Cardiovascular Epidemiology Training grant T32HL007024.
Duality of Interest. R.M.B. is a past volunteer of the American Diabetes Association. No other potential conflicts of interest relevant to this article were reported.
Author Contributions. E.S. designed the study and wrote the manuscript. E.S., C.M.P., and N.D. analyzed and interpreted the data. C.M.P., N.D., and R.M.B. reviewed and edited the manuscript. E.S. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
References
- 1.Selvin E, Parrinello CM, Sacks DB, Coresh J. Trends in prevalence and control of diabetes in the United States, 1988-1994 and 1999-2010. Ann Intern Med 2014;160:517–525 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ali MK, Bullard KM, Saaddine JB, Cowie CC, Imperatore G, Gregg EW. Achievement of goals in U.S. diabetes care, 1999-2010. N Engl J Med 2013;368:1613–1624 [DOI] [PubMed] [Google Scholar]
- 3.Stark Casagrande S, Fradkin JE, Saydah SH, Rust KF, Cowie CC. The prevalence of meeting A1C, blood pressure, and LDL goals among people with diabetes, 1988-2010. Diabetes Care 2013;36:2271–2279 [DOI] [PMC free article] [PubMed] [Google Scholar]