The GBD 2021 comprehensively evaluated over 370 diseases and injuries across 204 countries and regions globally, delivering detailed data on incidence, death, DALYs, and their corresponding age-standardized rates (20, 21). All data were derived from publicly available databases established by national institutions, with numerous published studies integrated and subjected to rigorous quality control measures to ensure data accuracy and reliability (22, 23).

Total incidence and age-standardized incidence rates (ASIR) were evaluated using Bayesian meta-regression models, whereas total mortality and age-standardized mortality rates (ASDR) were predominantly estimated utilizing the Cause of Death Ensemble Model (CODEm). Disability weights quantify the severity of health loss or non-fatal disabilities, while DALYs reflect the total years lost due to health impairment, from the onset of disease to death, serving as a crucial indicator for evaluating disease burden (24, 25).

The Socio-Demographic Index (SDI) is a composite measure that reflects a country's level of development, incorporating per capita income, the average education level of individuals aged 15 and older, and total fertility rates among individuals under 25 years. This index is closely associated with health outcomes, with SDI values ranging from 0 to 1, where 0 signifies the lowest level of development and 1 denotes the highest level of development (26, 27).

Based on SDI values, the 204 countries and regions were categorized into five distinct groups: high SDI, medium-high SDI, medium SDI, medium-low SDI, and low SDI regions. Patients were stratified into twelve age groups: <5 years, 5–9 years, 10–14 years, 15–19 years, 20–24 years, 25–29 years, 30–34 years, 35–39 years, 40–44 years, 45–49 years, 50–69 years, and 70+ years.

In the GBD 2021 study, the definition of nutritional deficiencies is grounded in the codes for nutritional deficiencies outlined in the 10th edition of the International Classification of Diseases (ICD), encompassing protein-energy malnutrition (ICD-10 codes E40–E46.9, E64.0), iodine deficiency (E00–E02), vitamin A deficiency (E50–E50.9, E64.1), and dietary iron deficiency (D50–D50.9) (28).

We assessed 70 specific attributable risk factors identified in GBD 2021, including particulate pollution, extreme temperatures, lead exposure, smoking, secondhand smoke exposure, high consumption of red meat and sodium, low intake of fiber, fruits, and vegetables, as well as elevated fasting blood glucose levels. However, owing to limitations in data availability, the final attribution analysis predominantly concentrated on iron deficiency, vitamin A deficiency, child underweight, and child wasting.

It was not appropriate or possible to involve patients or the public in the design, or conduct, or reporting, or dissemination plans of our research.

We evaluated the global burden of nutritional deficiencies by analyzing annual incidence, total death, DALYs, and the corresponding ASRs. Age-standardized rates serve to mitigate the effects of disparate age distributions and demographic shifts across various regions, thereby ensuring the comparability of study metrics. Utilizing the ASIR, ASDR, and age-standardized DALYs, we calculated estimates of the annual percentage change (EAPC) to retrospectively analyze the trends in the burden of nutritional deficiencies over the past 32 years. In the equation y = α + βx, y denotes the log10 (ASR) value, while x signifies the year. The formula for calculating EAPC is given by EAPC = 100 ^* (10^β – 1). If the EAPC value and its 95% confidence interval (CI) exceed zero, this indicates an increasing trend in ASR; conversely, it signifies a decreasing trend (23).

Furthermore, to investigate the correlation between ASIR, ASDR, age-standardized DALY rates, and levels of social development, we computed the Pearson correlation coefficients alongside the SDI values. Statistical analyses were performed utilizing GraphPad Prism (version 10.0.0 for Windows), encompassing t-tests and analysis of variance (ANOVA) to compare incidence rates and disability-adjusted life years across diverse genders, age groups, and regions.

All hypothesis tests conducted in this study were two-tailed, with a significance threshold established at P < 0.05 (29).

From 1990 to 2021, the global burden of nutritional deficiencies exhibited a consistent decreasing trend. The ASIR decreased by 54.9% (from 17,112.55 per 100,000 in 1990 to 7,725.10 per 100,000 in 2021, EAPC = −2.52), the ASDR declined by 72.2% (from 10.90 per 100,000 in 1990 to 3.03 per 100,000 in 2021, EAPC = −4.41), and the age-standardized DALY rate decreased by 51.9% (from 1,367.15 per 100,000 in 1990 to 657.62 per 100,000 in 2021, EAPC = −2.52). Furthermore, the ASIR for males was generally higher than that for females, whereas the age-standardized DALY rate was elevated for females; however, the difference in ASDR between genders was not statistically significant (Table 1, Figure 1A).

At the regional level, the burden of nutritional deficiencies in areas classified with a medium SDI and above continued to decline, with the lowest burden observed in high SDI regions. In low SDI regions, the ASDR and age-standardized DALY rate for nutritional deficiencies witnessed a significant increase between 2010 and 2012, peaking at 36.12 per 100,000 and 3,033.25 per 100,000, respectively; however, a subsequent consistent decreasing trend was observed (Table 1, Figure 1B).

Regionally, although the burden of nutritional deficiencies in Sub-Saharan Africa has diminished relative to 1990, it continues to be higher than in other regions. The ASIR in this region stands at 18,022.35 per 100,000, the ASDR is recorded at 9.56 per 100,000, and the age-standardized DALY rate is measured at 1,120.50 per 100,000. According to the World Bank's regional income statistics, as income levels increase, nutritional deficiencies diminish; the ASIR, ASDR, and age-standardized DALY rates in low-income areas are 11 times, 11 times, and over 10 times higher than those in high-income regions, respectively (Table 1).

Among the 204 countries analyzed, Somalia, Niger, and Chad exhibit the highest ASIR, with values of 83,940.53, 49,780.97, and 36,706.18, respectively. The highest ASDR values are reported in Sierra Leone, South Sudan, and Somalia, at 39.82, 33.90, and 32.07, respectively. The highest age-standardized DALY rates are recorded in Sierra Leone (2,901.99), Mali (2,742.30), and South Sudan (2,426.55). Furthermore, India ranks highest in total incidence, total mortality, and DALYs (Supplementary Tables S1–6, Figures 1C, D).

We further assessed the trends in the SDI across 21 regions worldwide from 1990 to 2021 and explored the potential associations with trends in ASIR, ASDR, and age-standardized DALYs. The results demonstrated that SDI exhibited a significant negative correlation with the ASIR of nutritional deficiencies (correlation coefficient = −0.48, P < 0.01), the ASDR (correlation coefficient = −0.51, P < 0.01), and the age-standardized DALY rate (correlation coefficient = −0.85, P < 0.01) (Figure 1E).

Figure 1F illustrates the results of a subgroup analysis stratified by age. The results revealed that except for high SDI regions, the age-standardized DALY rate for nutritional deficiencies among the < 5 years age group was consistently the highest. Conversely, in high SDI regions, the age-standardized DALY rate was highest among the 70+ years age group.

In 2021, the age-standardized DALY rate attributable to all risk factors associated with nutritional deficiencies was recorded at 611.69 per 100,000. The primary risk factors identified globally and across different SDI regions included iron deficiency, vitamin A deficiency, child underweight, and child wasting. Among these factors, iron deficiency emerged as the predominant risk factor contributing to disabilities and mortality associated with nutritional deficiencies (Figure 1G).

From 1990 to 2021, the global burden of protein-energy malnutrition demonstrated a generally fluctuating decreasing trend. The most substantial reductions were observed in ASDR, which decreased from 9.44 per 100,000 in 1990 to 2.61 per 100,000 in 2021 (with an EAPC of −4.47, indicating a decline of 72.4%), and in the age-standardized DALY rate, which fell from 699.3 per 100,000 in 1990 to 172.38 per 100,000 in 2021 (with an EAPC of −4.64, reflecting a decline of 75.3%). In contrast, the ASIR exhibited considerable fluctuations, with a significant decline occurring primarily after 2014. Furthermore, no significant differences were observed in the ASDR and age-standardized DALY rates for protein-energy malnutrition between males and females; however, the ASIR for males was notably higher than that for females (Table 2, Figure 2A).

At the SDI's regional level, the ASIR in middle SDI and higher regions displayed an upward trend before 2014, followed by a marked decline thereafter. By 2021, the ASIR in middle SDI regions was the highest, whereas that in high SDI regions was the lowest. Notably, in low SDI regions, the ASDR and age-standardized DALY rate for protein-energy malnutrition experienced a significant increase from 2010 to 2012, peaking at values of 34.39 per 100,000 and 2,037.19 per 100,000, respectively, before continuing to decline thereafter (Table 2, Figure 2B).

Regionally, in 2021, South Asia reported the highest ASIR for protein-energy malnutrition at 2,044.84 per 100,000, while Sub-Saharan Africa recorded the highest ASDR and age-standardized DALY rates at 9.14 per 100,000 and 451.59 per 100,000, respectively. According to the regional income classifications established by the World Bank, the burden of protein-energy malnutrition has diminished in both the World Bank Lower Middle Income and World Bank Low Income regions compared to 1990. Surprisingly, in 2021, both the ASDR and age-standardized DALY rates for protein-energy malnutrition increased in the World Bank High Income and World Bank Upper Middle-Income regions, while the ASIR in the World Bank High-Income region experienced a slight rise as well (Table 2).

Among the 204 countries, the highest ASIR was reported in the United Arab Emirates (4,688.09), followed by Bulgaria (2,563.81) and India (2,226.25). In contrast, the highest ASDR was observed in Sierra Leone (38.98), South Sudan (32.77), and Somalia (30.89), while the highest age-standardized DALY rates were recorded in Sierra Leone (2,200.17), South Sudan (1,891.77), and Mali (1,533.51). Additionally, India recorded the highest total incidence and total DALY years globally, with 29,505,320.45 cases and 2,035,962.502 years. Meanwhile, Indonesia ranked first in terms of mortality, with 18,148.40 deaths attributed to protein-energy malnutrition (Supplementary Tables S7–12, Figures 2C, D).

Figure 2E illustrates the potential correlation between the trends in changes of the SDI and the ASIR, ASDR, and age-standardized DALYs of protein-energy malnutrition across 21 global regions from 1990 to 2021. The results indicate a significant negative correlation between the SDI and the ASIR of protein-energy malnutrition (correlation coefficient = −0.14, P < 0.01), ASDR (correlation coefficient = −0.51, P < 0.01), and age-standardized DALY rate (correlation coefficient = −0.73, P < 0.01). Figure 2F presents the results of subgroup analyses categorized by age groups. The findings indicate that except for high SDI regions, the age-standardized DALY rate for protein-energy malnutrition in the under-5 age group is consistently the highest. Conversely, in high SDI regions, the age-standardized DALY rate for the 70+ age group is the highest.

From 1990 to 2021, the global burden of iodine deficiency exhibited a consistent year-on-year decline. The ASIR decreased from 126.14 per 100,000 in 1990 to 105.93 per 100,000 in 2021, reflecting a reduction of 16.2%, with an EAPC of −0.40 (−0.51, −0.29). The decline in the age-standardized DALY rate is even more pronounced, decreasing from 46.19 per 100,000 in 1990 to 27.66 per 100,000 in 2021, representing a decrease of 40.1%, with an EAPC of −1.56 (−1.66, −1.45). Notably, the burden of iodine deficiency is more pronounced in females compared to males (Table 3, Figure 3A).

At the regional level, the burden of iodine deficiency has decreased to varying degrees across all SDI categories, with more significant reductions observed in low-middle SDI and low SDI regions. Nevertheless, in 2021, the ASIR for iodine deficiency in low SDI regions was 9.71 times higher than that in high SDI regions (Low SDI: 20.54 per 100,000; High SDI: 199.41 per 100,000), while the age-standardized DALY rate was more than 15.95 times higher (Low SDI: 4.25 per 100,000; High SDI: 67.8 per 100,000) (Table 3, Figure 3B).

Regionally, most areas experienced a decline in the burden of iodine deficiency compared to 1990; however, a slight increase was noted in Eastern Europe, with the ASIR rising from 11.38 per 100,000 in 1990 to 11.52 per 100,000 in 2021, and the age-standardized DALY rate increasing from 4.42 per 100,000 in 1990 to 4.56 per 100,000 in 2021. Furthermore, in 2021, the burden of iodine deficiency in South Asia remains the highest, with an ASIR of 206.99 per 100,000 and an age-standardized DALY rate of 62.86 per 100,000.

According to the World Bank's classification of regional income statistics, areas with lower income levels also bear a relatively heavier burden of iodine deficiency (Table 3). Among 204 countries, Somalia, the Democratic Republic of the Congo, and Djibouti rank as the top three in terms of the burden of iodine deficiency, with Somalia's ASIR reaching 725.91 per 100,000. Additionally, India has the highest total number of cases and DALY associated with iodine deficiency globally, with 3,339,819.66 cases and 953,546.19 years (Supplementary Tables S13–16, Figures 3C, D).

Figure 3E illustrates the potential association between the trends of changes in the SDI and the ASIR and age-standardized DALYs of iodine deficiency across 21 global regions from 1990 to 2021. The results indicate a significant negative correlation between SDI and the ASIR of iodine deficiency (correlation coefficient = −0.40, P < 0.01) and the age-standardized DALY rate (correlation coefficient = −0.62, P < 0.01). Figure 3F presents the results of subgroup analyses based on age groups. The findings demonstrate that the DALYs related to iodine deficiency are primarily concentrated in the population aged over 5 years in both global and various SDI regions. Notably, in comparison to 1990, there was a slight increase in iodine deficiency-related DALYs for the 50–69 age group in high-middle SDI regions in 2021.

From 1990 to 2021, the global burden of vitamin A deficiency demonstrated a consistent downward trend annually. The ASIR decreased from 15,309.38 per 100,000 in 1990 to 6,212.97 per 100,000 in 2021, indicating a reduction of 59.4%, with an EAPC of −2.94 (−3.09, −2.79). The decline in the age-standardized DALY rate was even more pronounced, decreasing from 32.56 per 100,000 in 1990 to 15.73 per 100,000 in 2021, a reduction of 51.7%, with an EAPC of −2.43 (−2.63, −2.24). Furthermore, the burden of vitamin A deficiency is significantly greater in males than in females (Table 4, Figure 4A).

At the regional level, the burden of vitamin A deficiency has diminished to varying degrees across all SDI categories, with reductions, particularly notable in the low-middle and low SDI regions. Nonetheless, in 2021, the ASIR of vitamin A deficiency in low SDI regions remained 33 times higher than that in high SDI regions (Low SDI: 533.02 per 100,000; High SDI: 17,588.14 per 100,000), while the age-standardized DALY rate was over 121.03 times higher (Low SDI: 0.3 per 100,000; High SDI: 36.31 per 100,000) (Table 4, Figure 4B).

Regionally, in comparison to 1990, the burden of vitamin A deficiency has decreased across all regions. In 2021, the burden of vitamin A deficiency was highest in the Sub-Saharan Africa region, with an ASIR of 16,981.65 per 100,000 and an age-standardized DALY rate of 32.73 per 100,000.

According to regional income statistics from the World Bank, lower-income areas also bear a relatively greater burden of vitamin A deficiency (Table 4). Among 204 countries, Somalia exhibits the highest burden of vitamin A deficiency, with an ASIR of 81,451.76 per 100,000 and an age-standardized DALY rate of 108.60 per 100,000. Additionally, India holds the highest total incidence and DALY years for vitamin A deficiency globally, with 112,700,475 cases and 321,890.3 years (Supplementary Tables S17–20, Figures 4C, D).

Figure 4E illustrates the potential association between the trends of changes in the SDI and the ASIR and age-standardized DALYs of vitamin A deficiency across 21 global regions from 1990 to 2021. The results indicate a significant negative correlation between SDI and the ASIR of vitamin A deficiency (correlation coefficient = −0.49, P < 0.01) and the age-standardized DALY rate (correlation coefficient = −0.88, P < 0.01). Figure 4F presents the results of subgroup analyses based on age groups. The findings indicate that the DALYs associated with vitamin A deficiency are primarily concentrated in the population aged under 15 years in both global and various SDI regions.

From 1990 to 2021, the global burden of iron deficiency exhibited a consistent downward trend each year. The age-standardized DALY rate decreased from 517.98 per 100,000 in 1990 to 423.74 per 100,000 in 2021, representing a decline of 18.2%, with an EAPC of −0.68 (−0.72, −0.64). The burden of iron deficiency is significantly greater in females than in males (Table 5, Figure 5A).

At the regional level, although the burden of iron deficiency has diminished to varying degrees across all SDI regions, in 2021, the age-standardized DALY rate in low SDI regions was still more than 10.63 times higher than that in high SDI regions (Low SDI: 71.16 per 100,000; High SDI: 756.7 per 100,000) (Table 5, Figure 5B). Regionally, in comparison to 1990, the burden of iron deficiency has decreased across all regions. In 2021, the burden of iron deficiency was highest in the South Asia region, with an age-standardized DALY rate of 885.49 per 100,000.

According to regional income classifications by the World Bank, the lower-middle income region had the highest age-standardized DALY rate (674.98 per 100,000), whereas the high-income region had the lowest (72.89 per 100,000) (Table 5). Among 204 countries, Yemen exhibited the highest burden of iron deficiency, with an age-standardized DALY rate of 1,405.23 per 100,000. India reported the highest total DALY years attributable to iron deficiency globally, reaching 12,022,677.85 years (Supplementary Tables S21, 22, Figures 5C, D).

Figure 5E illustrates the potential association between the trends of changes in the SDI and age-standardized DALY rates for iron deficiency across 21 global regions from 1990 to 2021. The results indicated a significant negative correlation between SDI and the age-standardized DALY rate for iron deficiency (correlation coefficient = −0.75, P < 0.01). Figure 5F presents the results of subgroup analyses based on age groups. The findings indicated that the DALYs associated with iron deficiency were relatively evenly distributed across all age groups globally and within various SDI regions.

In comparison to existing studies, this research possesses several notable advantages. First, leveraging high-quality evidence and data frameworks from the latest GBD 2021 study, we offer a comprehensive analysis of the global trends in the burden of nutritional deficiencies and their four subtypes from 1990 to 2021. This study emphasizes the epidemiological characteristics of nutritional deficiencies across sex, country, region, and age group. Second, our findings provide crucial additional insights into the potential impact of the COVID-19 pandemic on the global burden of nutritional deficiencies, extending, and updating previous research over time.

Nonetheless, this study is not without its limitations. Firstly, the study results are contingent upon aggregated data from the GBD study, with their accuracy reliant on the quality of reporting from various countries. This reliance may lead to a significant number of undiagnosed cases of nutritional deficiencies in certain regions, thereby potentially impacting the accuracy of our findings (53). Secondly, there may be potential data quality issues arising from challenges in accurately determining the number of deaths and DALYs attributable to specific causes, such as nutritional deficiencies.