The Surging Tide of Diabetes: Implications for Nephrology

Diabetes is a leading cause of end stage renal disease (ESRD) and earlier stages of chronic kidney disease (CKD) in the United States and worldwide. In a recent study, Menke et al demonstrated that the prevalence of diabetes in the US increased substantially from 1988 to 2012, reaching 12.3-14.3% overall and higher among older adults and racial and ethnic minorities.¹ These results portend further downstream increases in diabetic kidney disease (DKD) and other DKD complications, such as cardiovascular diseases, infections, and premature mortality. Therefore, strategies to prevent DKD, apply proven therapies to high-risk populations, and identify new strategies to prevent and treat DKD warrant renewed effort and public health priority.

WHAT DOES THIS IMPORTANT STUDY SHOW?

Menke et al analyzed National Health and Nutrition Examination Survey (NHANES) data to determine the prevalence of diabetes in the US and its trends over time.¹ Included were data from 26,415 participants from either NHANES III (1988-1994) or the continuous NHANES cycles spanning 1999-2012. Each participant was studied only once. Evaluating groups of NHANES participants by their period of enrollment generated a series of snapshots of diabetes in the US population over time.

Diabetes was defined by self-reported physician diagnosis of diabetes, fasting plasma glucose ≥126 mg/dL, or hemoglobin A1c ≥6.5% (hemoglobin A1c/fasting plasma glucose definition), consistent with American Diabetes Association (ADA) guidelines.² A broader definition that additionally included oral glucose tolerance test 2-hour glucose ≥200 mg/dL was also assessed. NHANES sampling weights were applied to generate estimates of diabetes prevalence that were generalizable to the total civilian, non-institutionalized population of the US, accounting for differences in age distribution over time.

In 2011-2012, the prevalence of diabetes in the US was 12.3% using the hemoglobin A1c/fasting plasma glucose definition or 14.3% using the broader definition.¹ The prevalence of diabetes was substantially higher among older adults (using the hemoglobin A1c/fasting plasma glucose definition, 4.5%, 16.2%, and 24.7% for ages 20-44, 45-64, and ≥65 years, respectively), racial and ethnic minorities (20.6% for black participants, 16.5% for Asian participants, 18.7% for all Hispanic participants, and 20.5% for Mexican American participants, compared with 9.5% for white participants), and participants with lower income (17.8%, 11.5%, and 8.0% in ascending income categories). Diabetes was undiagnosed 25.2% of the time, with higher proportions of undiagnosed diabetes for younger participants and racial and ethnic minorities.

Over time, the prevalence of diabetes using the hemoglobin A1c/fasting plasma glucose definition increased from 9.8% in 1988-1994 to 12.3% in 2011-2012. This trend was not strictly monotonic, perhaps due to imprecise prevalence estimates with moderate numbers of diabetes cases (606-955 per 2-year NHANES cycle). The largest increase occurred from 1988-1994 to 2007-2008 (9.8% to 12.5%), with subsequent prevalence estimates of 12.1% and 12.3% in 2009-2010 and 2011-2012, respectively. Statistical tests revealed a strong linear trend over time (p<0.001) with no significant deviation from linearity. Increases in diabetes prevalence were due to increases in diagnosed diabetes, with a relatively constant prevalence of undiagnosed diabetes over time. In absolute terms, the increases in diabetes prevalence were greatest for older adults, blacks, Mexican Americans, and participants with lower income.

The study is strong because it evaluates large numbers of participants representative of the US population using standardized, quantitative measurements. The resulting estimates of diabetes prevalence are therefore the best available for the US population. There is a possibility that increased diabetes screening and awareness contributed to an increase in self-reported diabetes over time, inflating the apparent increase in diabetes prevalence to some extent. Also, the ADA recommends confirmation of elevated fasting plasma glucose or hemoglobin A1c prior to diagnosing diabetes,² and repeat measurements were not available. Some participants considered to have diabetes may therefore have had transient laboratory abnormalities that did not truly reflect diabetes; accounting for persistence in elevated fasting plasma glucose and hemoglobin A1c would lower diabetes prevalence in all time periods but be unlikely to affect temporal trends.

HOW DOES THIS STUDY COMPARE WITH PRIOR STUDIES?

Over the last 35 years, obesity, diabetes, CKD, and ESRD have each increased in prevalence. The new data by Menke et al should be viewed in the context of these broader trends. From 1980 to 2000, the prevalence of obesity in the United States doubled to reach 30%.³ Since then, the prevalence of obesity has stabilized, with 34.9% of US adults now obese.⁴ Sedentary lifestyle and ready access to calories are major contributors to this trend.

The prevalence of diabetes has followed the increasing prevalence of obesity, with a lag period of approximately one decade. An increase in the US prevalence of diabetes from 1988 through 2010 has been previously reported.^5,6 In comparison, the study by Menke et al extends the duration of time over which diabetes trends were examined. The new data emphasize the ongoing increase in the prevalence of diabetes. There was a suggestion that the rate of increase in diabetes is slowing, consistent with a study demonstrating a leveling of self-reported diabetes prevalence over the same period;⁷ however, a test for deviation from a linear trend in diabetes prevalence was not significant. The study by Menke et al also highlights several subgroups with particularly high risk of diabetes, including older adults, African Americans, Asian Americans, and Hispanic Americans. Looking more broadly, 387 million people worldwide currently have diabetes with an expected rise to nearly 600 million over the next 20 years (through 2035).⁸ Most of them (77 %) live in low- and middle-income countries where access to treatments of any sort, especially advanced therapeutics like dialysis and transplantation, is often very limited.

These trends in diabetes directly link to trends in the prevalence of DKD. In the US, DKD increased in proportion to the prevalence of diabetes from 1988 through 2008.⁹ It follows that the increased prevalence of diabetes documented by Menke et al has been or will be followed by an increase in the prevalence of DKD. Moreover, the US population is aging, and racial and ethnic diversity is increasing. Older adults and racial and ethnic minorities are at high risk of CKD and ESRD.⁸ The high prevalence of diabetes in these subgroups suggests that older adults and racial/ethnic minorities will drive the increase in incidence of DKD.

Diabetes is the most common cause of ESRD in the US, accounting for 50,534 cases of incident ESRD (44%) in 2012.⁸ Incidence rates of ESRD attributed to diabetes increased from 11 to 143 per million from 1980 to 2000,⁸ and, since 2000, incidence rates of ESRD attributed to diabetes have plateaued. As the prevalence of earlier stages of DKD has increased over the same period, it is possible that stabilization of incident ESRD attributed to diabetes is due to reduced progression of DKD to ESRD associated with better diabetes care, including the widespread use of renin-angiotensin system inhibitors for control of blood pressure and albuminuria.⁹ However, the increasing prevalence of diabetes reported by Menke et al, and the proportional increase in DKD previously noted, provide an ominous warning that there is a growing population at risk of ESRD and earlier stages of CKD due to diabetes and related complications, such as cardiovascular diseases, infections, and premature mortality.^10,11 This public health threat is worthy of serious and focused attention.

WHAT SHOULD CLINICIANS AND RESEARCHERS DO?

The increasing prevalence of diabetes has multiple important implications for the field of nephrology. First and foremost, the increase in diabetes prevalence suggests that the public health impact of DKD will continue to grow. However, changes in diabetes screening, prevention, and treatment could potentially blunt the potential rise in DKD. To better ascertain public health priorities, further data on the prevalence and incidence of DKD and related complications are needed. In the US, particular attention should be paid to high-risk subpopulations of older adults and racial and ethnic minorities. Analysis of NHANES data and ongoing follow-up through the United States Renal Data System are critical for this purpose, but better estimates of the worldwide burden of ESRD and earlier stages of CKD due to diabetes are also required. These data will inform planning for payment systems and workforce and help prioritize allocation of research funding for new prevention and treatment strategies.

In addition, the new data from Menke et al suggest that clinicians should enhance screening for and implementation of the recommended standards of care for diabetes and DKD, particularly in high-risk populations. The ADA, National Kidney Foundation–Kidney Disease Outcomes Quality Initiative (NKF-KDOQI), and Kidney Disease: Improving Global Outcomes (KDIGO) recommend that patients with diabetes be screened for DKD yearly with urine albumin-creatinine ratio and estimated GFR, starting after 5 years of type 1 diabetes and at the time of diagnosis of type 2 diabetes.¹² Currently, most professional guidelines recommend treating hypertension to a goal blood pressure of <140/90 mm Hg, with lower targets considered for some patients, such as those with higher albuminuria.¹² Renin-angiotensin system inhibitors prevent progression of DKD and reduce risk of cardiovascular events. Effective implementation of screening recommendations and proven interventions is critical. Such implementation is currently suboptimal in racial and ethnic minorities,¹³ and implementation strategies should be designed to target these high-risk groups.

Finally, the new data emphasize an imperative to continue searching for novel approaches to prevent and treat diabetes and DKD. Treatments to slow progression of DKD as well as the cardiovascular complications of diabetes and DKD have enjoyed a renewed focus of investigation.¹⁴ The importance of these directions is reinforced by the new data. In addition, given the sheer magnitude of the global burden of diabetes and DKD, additional emphasis must be placed on accessible, safe, and inexpensive interventions that can be widely disseminated and implemented to prevent diabetes and DKD.