Global, Regional, and National Temporal Trends in Incidence for Type 2 Diabetes Mellitus Related Chronic Kidney Disease from 1992 to 2021

Article information

Diabetes Metab J. 2025;.dmj.2024.0593

Publication date (electronic) : 2025 March 11

doi : https://doi.org/10.4093/dmj.2024.0593

Yu Cao^1,2, Huiting Chen^1,2, Hui Liu^1,2, Hao Wu^1,2, Wei Gao^,1,2

¹School of Public Health, Jiangxi Medical College, Nanchang University, Nanchang, China

²Jiangxi Provincial Key Laboratory of Disease Prevention and Public Health, Nanchang University, Nanchang, China

Corresponding author: Wei Gao https://orcid.org/0000-0001-8298-3415 School of Public Health, Jiangxi Medical College, Nanchang University, 461 Bayi Ave, Nanchang, Jiangxi 330006, China E-mail: wei.gao@ncu.edu.cn

Received 2024 September 26; Accepted 2024 November 21.

Abstract

GRAPHICAL ABSTRACT

Highlights

INTRODUCTION

Chronic kidney disease (CKD) is a chronic, progressive condition of the kidney, lasting 3 months or more, characterized by a loss of the kidney’s key function to filter blood and produce urine, potentially leading to the need for renal replacement therapy [1]. CKD has emerged as a significant global public health concern. According to the Global Burden of Disease (GBD) 2021 study, the global prevalence of CKD at 8.53% in 2021, resulting in a disability-adjusted-life-years of 1.54% of the total burden, and ranked as the 20th leading cause of disease [2]. Most people with CKD live in areas with poor primary healthcare infrastructure, lack access to diagnostic treatment, and having low disease awareness, leading to a lack of active treatment-seeking [3]. Type 2 diabetes mellitus (T2DM)-related CKD (CKD-T2DM) is the major subtype of CKD, carrying the highest disease burden among all subtypes of CKD with a clear etiology [4]. The CKD often co-occurs with major non-communicable diseases such as ischemic heart disease, stroke, and cancer [3], and is also interlinked with acute kidney injury, which increases the risk of acute kidney injury and results in serious short-term clinical outcomes [5].

Population aging and population growth will be the theme of world demographic trends in the coming decades [6]. The disease burden of T2DM will also continue to increase, with 1.3 billion people expected to be affected by 2050 [7]. Changes in macro-factors such as population growth, aging, and the increasing burden of T2DM are key drivers of CKD incidence and have likely influenced CKD incidence trends. To date, previous studies on global trends in CKD-T2DM incidence have been mainly descriptive or focused on specific populations [8,9], and comprehensive analyses have not been fully explored. Therefore, a detailed analysis of global temporal trends in CKD-T2DM incidence is essential for monitoring progress and guiding health investment priorities. This analysis should consider the impacts of demographic changes and population growth on CKD-T2DM incidence. Additionally, focusing on the relationship between incidence trend and age, period, and cohort effects can evaluate healthcare delivery success and identify remaining regional disparities. These integrated approaches provide a unique opportunity to explore the burden of CKD-T2DM incidence at global, regional, and national levels.

The purpose of this study is to update previous global CKD-T2DM incidence studies using the latest GBD 2021 data, characterizing CKD-T2DM incidence from 1992-2021. Decomposition methods and age-period-cohort (APC) modeling were then used to examine the impact of macro-factors such as population, age, period, and cohort on the incidence of CKD-T2DM over the last 30 years. The results will help identify spatial and temporal trends, elucidate changes in disease patterns, and provide guidance for the development of prevention strategies and management measures to reduce the burden of CKD-T2DM.

METHODS

RESULTS

Global, regional, and national trends in CKD-T2DM incidence, 1992−2021

Fig. 1, Supplementary Tables 4 and 5 show the total number of incidents, all-age incidence, ASIR, and net drift of ASIR at global, regional, and national level in 1992 and 2021. From 1992 to 2021, the global incidents of CKD-T2DM increased from 801.87 (95% UI, 725.55 to 878.5) thousand cases to 2,012.02 (95% UI, 1,857.8 to 2,154.29) thousand cases, the all-age incidence from 14.59 (95% UI, 13.2 to 15.98) per 100,000 people to 25.5 (95% UI, 23.54 to 27.3) per 100,000 people, while ASIR from 19.39 (95% UI, 17.59 to 21.16) per 100,000 people to 23.07 (95% UI, 21.4 to 24.72) per 100,000 people. Globally, the APC model estimated a net drift of 0.4% (95% confidence interval [CI], 0.32 to 0.47) in the CKD-T2DM ASIR from 1992 to 2021 (Supplementary Table 4).

Fig. 1.

The all-age incidence in 1992 (A) and 2021 (B) for type 2 diabetes mellitus related chronic kidney disease in 204 countries and territories.

Regionally, CKD-T2DM incidents, all-age incidence, and ASIR have increased in all regions over the last 30 years. The most significant increases were observed in the middle SDI regions, with increases of 224.72%, 136.03%, and 31.63%, respectively. In 2021, the burden of incidence remained highest in the high SDI region, with all-age incidence and ASIR of 54.41 (95% UI, 50.08 to 58.21) and 28.34 (95% UI, 26.19 to 30.3) per 100,000 people. According to the net drift results, the middle SDI region has the highest annual growth rate of ASIR at 0.69% (95% CI, 0.61 to 0.76), while the high SDI region has the lowest annual growth rate of ASIR at 0.13% (95% CI, –0.02 to 0.28) (Supplementary Table 4).

At the national level, there were 170 countries and territories with more than 100 cases of CKD-T2DM in 2021. Among them, China (354.16 thousands; 95% UI, 321.26 to 382.78), India (222.79 thousands, 95% UI, 199.08 to 245.04), the United States (198.28 thousands; 95% UI, 180.73 to 216.56), and Japan (105.93 thousands; 95% UI, 95.55 to 115.89) accounted for 68.7% of the global CKD-T2DM incidents. In 2021, the countries with the highest and lowest all-age incidence were Japan 83 (95% UI, 74.8 to 90.8) per 100,000 people and Somalia 2.9 (95% UI, 2.5 to 3.4) per 100,000 people. The highest ASIR was in Qatar 53.1 (95% UI, 46.3 to 60.1) per 100,000 people and the lowest in Madagascar 10.3 (95% UI, 8.9 to 11.7) per 100,000 people. Among these 170 countries, the highest increases in incidence and ASIR are Albania (330.59%) and Estonia (108.05%), respectively, and only five countries (Afghanistan, Chad, Guinea, Liberia, and Mozambique) show a decline in all-age incidence, all of which were low SDI countries; five countries (Ireland, Greece, Italy, Spain, and Costa Rica) showed a decrease trend in ASIR. Overall, national trends in CKD-T2DM incidence were consistent with SDI regional trends, but varied across countries. The net drift changed from –1.05% (95% CI, –2.57 to 0.5) in Ireland to 2.23% (95% CI, –3.81 to 8.65) in El Salvador. Some countries exhibited discrepancies between incidence changes (all-age/age-standardized) and net drift from the APC model, suggesting that the change in CKD-T2DM incidence over the 30 years was not entirely a linear trend, highlighting the need to distinguish between period and cohort trends in CKD-T2DM (Fig. 1, Supplementary Table 5).

Drivers of CKD-T2DM epidemiology: population growth, aging, and epidemiological changes

Fig. 2 illustrates the association between the all-age incidence, ASIR of CKD-T2DM and SDI. A strong positive correlation (R=0.795, P<0.001) was found between all-age incidence of CKD-T2DM and SDI (Fig. 2A). The incidence increasing sequentially across SDI quintile, from 6.92 (95% UI, 6.22 to 7.6) per 100,000 people in low SDI regions to 54.41 (95% UI, 50.08 to 58.21) per 100,000 people in high SDI regions. However, the ASIR of CKD-T2DM did not exhibit a strong correlation with SDI (R=0.423, P<0.001) after adjusting for demographic structure (Fig. 2B). Population changes are associated with variations in incidence burden, and decomposition analyses were used to explore these relationships.

Fig. 2.

Relationship between all-age incidence rate (A) and age-standardized incidence rate (B) of type 2 diabetes mellitus related chronic kidney disease and sociodemographic index (SDI). Each circle represents a country; circles are colored according to SDI quintile; circle size corresponds to incidents number. The top 10 countries in terms of incidents are labeled.

Fig. 3 depicts the decomposition of the increase in CKD-T2DM incidents from 1992 to 2021, attributing changes to population aging, changes in population size, and changes in incidence (termed epidemiological changes). Globally, population aging, population growth, and epidemiological changes contributed 31.4%, 49.2%, and 19.3%, respectively, to the overall increase in CKD-T2DM incidents between 1992 and 2021. In high SDI regions, aging had the greatest impact on total incidents (47%); population growth predominated in low SDI regions (87%); and epidemiological changes were most influential in middle SDI regions (27.6%). Supplementary Table 6 details the incident decomposition for each country.

Fig. 3.

Changes in type 2 diabetes mellitus related chronic kidney disease incidents according to population-level determinants of aging, population growth, and epidemiological change from 1992 to 2021 at the global level and by sociodemographic index (SDI) regions. Black dots represent the overall value of change contributed by all three components.

Time trends in CKD-T2DM incidence rate across different age groups

Fig. 4A shows the annual percentage change in CKD-T2DM incidence across 5-year age groups from 15 to 94 years. Globally, a decreasing trend in CKD-T2DM incidence is observed in the 15–29 years age group (local drift <0), while an increasing trend is noted in the 30–94 years age group (local drift >0). The increase in incidence was greater in the 40–79 years age range, local drift was higher than 0.5%. As age continued to increase, the upward trend diminished, from 0.59 (95% CI, 0.52 to 0.65) in the 75–79 years age group to 0.07 (95% CI, –0.29 to 0.44) in the 90–94 years age group. Gender differences show slightly higher local drift for females compared to males, but this distinction is not statistically significant. Similar patterns are observed across all SDI regions. Supplementary Figs. 1-5 detail the local drift of incidence for each country.

Fig. 4.

Local drifts of type 2 diabetes mellitus related chronic kidney disease (CKD-T2DM) incidence (A) and age distribution of incidents (B) from CKD-T2DM by sociodemographic index (SDI) quintiles, 1992−2021. (A) Local drifts of CKD-T2DM incidence (estimates from age-period-cohort models) for 16 age groups (15−19 to 90−94 years), 1992−2021. The dots and shaded areas indicate the annual percentage change of incidence (% per year) and the corresponding 95% confidence intervals. (B) Temporal change in the relative proportion of CKD-T2DM incidents across age groups (15−39, 40−49, 50−59, 60−69, 70−79, and 80−94 years), 1992−2021.

Fig. 4B illustrates the temporal trends in the incidents of CKD-T2DM by age. Globally, the incidents in the elderly population (60–94 years) are increasing as a proportion of the total incidents, with 75% of CKD-T2DM cases worldwide occurring in people aged 60 years and older in 2021, and this proportion reaches 86.93% in high SDI regions. Within all regions, the elderly population constitutes the primary demographic affected by CKD-T2DM. However, in middle, low-middle, and low SDI regions, the incidents proportion in the oldest age group (>80 years) remains relatively low. Supplementary Figs. 6-10 provide detailed age distributions of incidence for each country.

Age, period, and cohort effects on CKD-T2DM incidence

Fig. 5 illustrates the age, period, and cohort effects on CKD-T2DM incidence. Generally, the age effect exhibits a similar pattern across SDI regions, with incidence risk increasing with age, peaking around 75–84 years, and then declining, indicating a high risk among the elderly. In high SDI regions, the incidence risk is higher in older age groups compared to other regions. Although males show a higher age effect than females in older age groups, this difference was not significant (Fig. 5A).

Fig. 5.

Age, period and cohort effects on type 2 diabetes mellitus related chronic kidney disease incidence by sociodemographic index (SDI) quintiles. (A) Age effects are shown by the fitted longitudinal age curves of incidence (per 100,000 person-years) adjusted for period deviations. (B) Period effects are shown by the relative risk of incidence (incidence rate ratio) and computed as the ratio of age-specific rates from 2002 to 2006 (the referent period) to 2017−2021. (C) Cohort effects are shown by the relative risk of incidence and computed as the ratio of age-specific rates from the 1898 cohort to the 2006 cohort, with the referent cohort set at 1987. The dots and shaded areas denote incidence rates rate ratios and their corresponding 95% confidence intervals.

Globally, the period effect shows an increased risk of incidence over the past 30 years. In high SDI regions, the period effect for incidence risk remains relatively stable. However, in high-middle and middle SDI regions, the period effect continues to rise, indicating that CKD-T2DM incidence has not been effectively controlled. The period effect in low-middle and low SDI regions show a rapid increase in incidence risk over the last 15 years (Fig. 5B).

Globally, the cohort effect was observed within each SDI region to show a rising and then falling trend in consecutive birth cohorts. The rising cohort effect is more pronounced in the middle-high and middle SDI regions. The risk across regions begins to decline around the 1977 cohort (Fig. 5C). Supplementary Figs. 11-25 detail the effects of age, period, and cohort on CKD-T2DM incidence in each country.

Age, period, and cohort effects in exemplary countries

To characterize the significant changes in CKD-T2DM incidence by APC effects globally, we selected representative countries across SDI quintiles. Fig. 6A highlights countries with favorable trends. Japan shows notable period and cohort effects with a significant downward trend in ASIR of CKD-T2DM. Italy exhibits the best trend among high-middle SDI countries, with a net drift of –0.72% (95% CI, –1.44 to 0) and declining period and cohort effects. Costa Rica has a declining ASIR, though its period and cohort effects remain relatively stable. In India, individuals over 60 years old are at high risk, with local drifts exceeding 0%, but the overall increase in incidence is not significant, accompanied by declining period and cohort risks throughout the study. However, few low SDI countries showed favorable trends, with Somalia performing relatively well, with only a 14% increase in ASIR and relatively stable local drift for those in the age groups over 35.

Fig. 6.

Favorable (A) and unfavorable (B) age-period-cohort effects on exemplar countries across sociodemographic index (SDI) quintiles. Age distribution of incidents shows the relative proportion of incidents from each age group during 1992−2021. Local drifts indicate the annual percentage change of incidence (% per year) across 5-year age groups (from 15−19 to 90−94 years). Age effects are represented by the fitted longitudinal age curves of incidence (per 100,000 person-years) adjusted for period deviations. Period effects are shown by the relative risk of incidence (incidence rate ratio) and computed as the ratio of age-specific rates from 2002 to 2006 (the referent period) to 2017−2021. Cohort effects are shown by the relative risk of incidence and computed as the ratio of age-specific rates from the 1898 cohort to the 2006 cohort, with the referent cohort set at 1987. The dots and shaded areas denote incidence rates rate ratios and their corresponding 95% confidence intervals.

In contrast, Fig. 6B illustrates countries facing unfavorable APC effects, indicating an increased risk of incidence. Unlike many high SDI countries, Saudi Arabia has experienced a dramatic rise in incidence, with a 4-fold increase over the past 30 years and a 51.3% increase in ASIR, alongside rising period and cohort effects. China has garnered attention for its substantial number of CKD-T2DM incidents, which continue to rise across all-age groups over 35, with increasing period and cohort effects. Mexico exemplifies many middle SDI countries, displaying rising incidence across all-age groups and escalating period and cohort effects. Kenya and Ethiopia represent underdeveloped nations, while their CKD-T2DM incidence are not high, both have seen ASIR increases of over 20% in the last 30 years, accompanied by significant growth in period and cohort effects.

DISCUSSION

Diabetes is a major independent risk factor for CKD, causing about 10% of CKD and 40% of end stage renal disease (ESRD) cases [2,15], CKD-T2DM has become a major public health problem. Our study found that the burden of CKD-T2DM rapidly increased from 1992 to 2021, with increasing global health disparities and a gradual increase in incidence from low to high SDI regions. This rapid progression is closely linked to the onset and acceleration of T2DM globally [16]. Some studies suggest that the COVID-19 pandemic has heightened the risk of developing T2DM and CKD [17,18]. However, our findings indicate that during the pandemic, the incidents and all-age incidence continued to rise at a consistent pace without a marked increase. The possible explanation is that T2DM leading to CKD is a long-term process and the short-term effect is not apparent [19], and more cohort studies are needed to demonstrate the association between COVID-19 and CKD-T2DM.

Between 1992 and 2021, the global population grew by 43.55%, but CKD-T2DM incidents, incidence, and ASIR rose by 150.92%, 74.79%, and 18.98%, respectively. The unbalanced population, incidence, and ASIR growth trends are of concern to us, and the results of the decomposition analysis provide an explanation. First, aging and population growth are key drivers of incidents growth in high and low SDI regions, respectively, and the key drivers in different regions correspond to local population development patterns. High SDI areas have completed or are experiencing population aging, whereas low SDI areas have very rapid population growth and relatively young demographics [20]. Second, the underlying epidemiologic trend (incidence of CKD-T2DM) contributes about one-fifth of the incidents growth across regions, suggesting opportunities to reduce the CKD-T2DM burden despite inevitable population growth and aging. Addressing CKD-T2DM requires focusing on preventing T2DM and its progression to CKD. We analyzed global, regional, and national CKD-T2DM patterns using the APC model, and the trends in various areas can provide valuable lessons for mitigating CKD-T2DM incidence growth.

In all regions, the elderly are the primary population for CKD-T2DM incidence, with a significant proportion of cases and a continuously increasing incidence (local drift ≥0). The risk of onset of CKD-T2DM peaked around age 80, occurring earlier in low and middle SDI regions. This pattern mirrors T2DM, where the risk of T2DM onset peaks around age 60 and the risk of CKD onset follows due to a lag effect [16,19]. Lower life expectancy in low and middle SDI regions may lead to premature death before CKD can develop [11]. Notably, approximately 2% of CKD patients progress to ESRD, with older patients at higher risk [21]. Patients with ESRD often require dialysis or kidney transplantation, but there is a significant imbalance in the global capacity to cope with ESRD, with patients treated for ESRD residing predominantly in high-income countries that have affordable access to healthcare [22]. Additionally, CKD is often asymptomatic until late stages, and outcomes are worse in older patients, making early prevention and treatment critical.

Favorable period effects are rarely observed, mainly in high and middle-high SDI countries such as Japan, Republic of Korea, Italy, and Spain. Conversely, unfavorable period effects are common in many regions. Japan, burdened by aging and diabetes, has seen an increase in diabetes and CKD patients [23]. Since 2000, Japan’s Health Japan 21 policy and the 2016 Diabetic Kidney Disease Prevention Program were initiated, and efforts have been made to ensure the quality of intra-regional cooperation to address these issues [23,24]. The Korean government’s initiative to offer a free, regular health checkup program for individuals over 40, along with the increased use of antidiabetic medications, may have contributed to a reduction in CKD incidence [25-27]. Countries like the United States, the United Kingdom, Germany, and Singapore show stable period effects, likely due to advanced local CKD surveillance, diagnostic, and treatment technologies [28]. Period effects in places like China and Ethiopia increase the risk of CKD-T2DM incidence, but the reasons may be inconsistent. China has invested heavily in public health surveillance, including the expansion of the China Chronic Disease and Risk Factor Surveillance system from 79 sites in 2004 to 302 sites in 2021, has improved case detection [29,30]. Despite increased incidence risk suggested by period effects, better surveillance in China is beneficial. Effective monitoring could improve the use of healthcare services, aiding in the control of CKD and its comorbidities and ultimately reducing the disease burden [31, 32]. In contrast, World Health Organization guidelines and packages for primary care interventions for T2DM prevention and treatment are underutilized in sub-Saharan Africa due to resource constraints and lack of strong action [33].

Cohort effects indicate an upward trend in CKD-T2DM incidence risk across areas, especially in middle and low SDI areas. Although the incidence risk appears to decline starting with the 1982 cohort, this is because these cohorts haven’t yet reached the high-risk age. In recent decades, rapid regional economic development and urbanization have led to substantial shifts in lifestyle and diet [34]. Fast-paced modern lifestyles are often characterized by high-sugar, high-fat diets and increased sedentary behaviors, and this lifestyle shift is contributing to the global increase in overweight and obesity, which in turn increases the risk of diabetes and its complications [15,35,36]. Particularly in countries representative of the middle and low SDI countries, such as China, Nigeria, and Mexico have experienced rapid urbanization and dietary shifts [37]. Birth cohorts undergoing these changes haven’t yet reached high-risk ages for T2DM, but their incidence risk is likely substantial.

The APC model results emphasize the importance of improving the risk of developing CKD-T2DM at the primary care level and in terms of healthy living. Facility-based kidney disease prevention, identification, and treatment programs are cost-effective strategies, but CKD screening has lagged due to the persistent lack of effective therapies. This has changed with new treatments like sodium glucose cotransporter 2 (SGLT2) inhibitors, glucagon like peptide-1 receptor agonists, endothelin receptor antagonists, and selective mineralocorticoid receptor antagonists [38], means that early identification and treatment of CKD can translate into large-scale health improvements. For instance, a 2021 meta-analysis showed that SGLT2 inhibitors reduced CKD progression risk by 37% in diabetics [39]. In addition, independent screening for CKD is difficult to achieve due to the limited availability and public funding for CKD care, particularly in low SDI regions (sub-Saharan Africa, South Asia, Middle East, etc.). Considering the integration of CKD screening with cardiovascular or diabetes disease prevention efforts may therefore be an opportunity to address global inequalities in kidney disease care [40]. Diabetes prevention trials have shown that maintaining normoglycemia long-term is possible in high-risk populations [41]. Implementing disease management programs promoting physical activity, healthy diets, and weight loss can help prevent or delay CKD-T2DM progression.

To the best of our knowledge, this is a comprehensive effort to utilize GBD2021 data to analyze the global, regional, and national burden of CKD-T2DM incidence, assessing current status, long-term trends, and regional differences. Our findings support cross-sectional comparisons between different regions and countries and update data beyond 2020, which is an advancement compared to previous studies [42,43]. In addition, this study analyses and captures in detail the independent effects of macro-factors such as demographic change, age, period and cohort on trends in CKD-T2DM incidence, thus capturing important trends in specific populations and highlighting timely success points and potential key areas. This study also has limitations. High-quality population-based studies on CKD-T2DM incidence are lacking in many countries, and the GBD primarily relies on statistical methods and predictive covariates, which can compromise data accuracy, particularly in low and middle SDI regions. This may lead to overestimation of the effect values derived from the APC model. Second, the 5-year age group incidence used in this study, which is the predominant age-interval data format in the APC model, had partially overlapping birth cohorts, complicating identification of the youngest age cohort (<1 year). Third, the GBD assumes all diabetics under 15 years have type 1 diabetes mellitus, excluding younger populations from this study despite evidence of T2DM in children [44]. Future research should aim to obtain raw data on CKD-T2DM incidence in low and middle SDI countries, assess CKD-T2DM incidence in children, and perform APC modeling on younger age groups to reveal more valuable insights. In addition, this study used cross-sectional data, and future more longitudinal individual studies are needed to supplement and validate the period and cohort risk changes found in the study.

In conclusion, globally, the incidence cases of CKD-T2DM have risen dramatically over the past 30 years, driven largely by population growth and aging. While CKD-T2DM incidence are highest in high SDI regions, they are rising rapidly in lower SDI countries. Unfavorable period and cohort effects suggests inadequate resources for CKD screening and care in many developing nations. Rapidly developing areas, with their evolving lifestyles, are also at heightened risk. Given its largely preventable and treatable nature, CKD-T2DM demands greater attention in global health policies. Implementing integrated CKD and diabetes screening programs, particularly targeting older populations, can significantly reduce T2DM complications, slow CKD progression, and improve health outcomes.

SUPPLEMENTARY MATERIALS

Article information Continued