Burden of micronutrient deficiency among patients with type 2 diabetes: systematic review and meta-analysis

Type 2 diabetes mellitus (T2D) is a growing global health problem driven by insulin resistance, impaired insulin secretion, obesity, sedentary lifestyles, and diet. Micronutrients act as cofactors in glucose metabolism, pancreatic beta-cell function, and insulin signalling; deficiencies may exacerbate insulin resistance through oxidative stress and impaired enzyme activity. Prior studies report varied and sometimes contradictory estimates of individual micronutrient deficiencies (for example, widely differing rates of vitamin D deficiency) among people with T2D, complicating clinical and policy decisions. This review asks: What is the burden (prevalence) of micronutrient deficiencies among patients with T2D? The purpose is to synthesise global evidence on multiple micronutrients to inform practice and policy and to establish a baseline for future research.

Background literature indicates that approximately one-third of the global population has at least one essential micronutrient deficiency, and deficiencies in nutrients such as biotin, chromium, thiamine, vitamin D and vitamin C have been linked to insulin resistance and T2D pathophysiology. Prior systematic reviews reported widely varying prevalence for vitamin D deficiency in T2D (for example, 32.7% vs 80.4%), and numerous primary studies have focused on single micronutrients with inconsistent findings. This review addresses these inconsistencies by aggregating data across multiple micronutrients, geographies, and time periods.

Design: Systematic review and meta-analysis conducted per PRISMA 2020 and the Cochrane Handbook; protocol published previously. Search strategy: Comprehensive searches in Embase, ProQuest, PubMed, Scopus, Cochrane Library, Google Scholar, LILACS, and grey literature (including institutional repositories such as the International Diabetes Federation). Initial search to 31 July 2023, updated 1 January 2024 to include studies through 31 December 2023. Keywords combined MeSH/Emtree and free-text terms related to “micronutrient deficiency” and “type 2 diabetes” and “prevalence/incidence/burden.” Search strategy peer reviewed using the PRESS checklist. Eligibility criteria: Observational studies (cross-sectional, longitudinal, cohort) and RCTs reporting prevalence/burden/incidence of deficiency of minerals/electrolytes (zinc, chromium, iron, copper, fluoride, selenium, iodine, manganese, calcium, phosphorus, molybdenum, potassium, folic acid, magnesium, sodium) and vitamins (A, B1, B2, B3, B5, B6, B7, B9, B12, C, D, E, K) among adults (≥18 years) with T2D, with or without complications, all sexes and ethnicities, published 1998–2023 without language or geographic restriction, plus grey literature. Excluded: T1D, gestational diabetes, T2D <18 years, supplementation trials for deficiency estimates, case reports/series, narrative reviews, editorials, conference abstracts/posters, ecological studies. Study selection: Records deduplicated and screened in Rayyan by independent reviewers (titles/abstracts: NS, KCS; full text: NS, DKM) with disagreements resolved by a third reviewer (SDG). Non-English articles translated via Google Translate. Authors were contacted for unavailable full texts; studies excluded after three unsuccessful follow-ups. Reasons for exclusion documented; cross-references checked. Data extraction: Two reviewers (NS, DG) independently extracted study characteristics, micronutrients assessed, T2D and deficiency diagnostic criteria and cut-offs (harmonized across measurement units where possible). Discrepancies resolved with a third reviewer (DKM). Risk of bias/quality assessment: Joanna Briggs Institute (JBI) Critical Appraisal tools applied according to study design (prevalence, analytical cross-sectional, SRMA, cohort). Studies graded for risk (low/unclear/high) and used in sensitivity analyses. Statistical analysis: Meta-analyses of proportions using random-effects models. Variance-stabilizing approaches evaluated: no transformation, logit, double arcsine (Freeman–Tukey), and generalized linear models; model fit compared via AIC/BIC. Pooled prevalence estimated overall (any micronutrient deficiency) and for micronutrients with ≥3 studies (vitamin D, vitamin B12, magnesium, iron). Weights were inverse-variance; SEs and 95% CIs derived from counts where not reported. Heterogeneity quantified via τ² and I²; publication bias assessed by Egger’s regression and funnel plots; trim-and-fill applied. Subgroup analyses: By sex, diabetic complications, study setting (hospital vs community), WHO region (Americas, Europe, Eastern Mediterranean, South-East Asia, Western Pacific, Africa), metformin exposure for vitamin B12, and time periods (1998–2015 vs 2016–2023). Meta-regression: Publication year as covariate to assess temporal trends overall and for individual micronutrients; visualized with bubble plots. Sensitivity analyses: (1) Excluding studies with weight >10%; (2) Excluding outliers identified via standardized residuals. Analyses conducted in R v4.3.2 using tidyverse, meta, and metafor.

- Included 132 datasets (from 127 studies) with 52,501 participants. - Overall pooled prevalence of any micronutrient deficiency among T2D patients: 45.30% (95% CI 40.35% to 50.30%) using random-effects with double arcsine transformation; heterogeneity I²≈99% (p<0.001). Sensitivity excluding high-weight studies yielded 47% (95% CI 42% to 51%). Egger’s test indicated funnel plot asymmetry (z=2.0269, p=0.0427); trim-and-fill limit estimate 0.6211 (95% CI 0.4973 to 0.7450). - By sex: Men 42.53% (95% CI 36.34% to 48.72%; k=62); Women 48.62% (95% CI 42.55% to 54.70%; k=63); significant heterogeneity in both. - Diabetic complications subgroup (k=26): 39.83% (95% CI 29.38% to 50.28%). - Setting: Hospital-based (k=124; n=43,367): 46% (95% CI 41% to 51%); Community-based (k=5; n=8,839): 22% (95% CI 6% to 46%). - WHO regions pooled prevalence: Americas 54.04% (95% CI 35.03% to 72.48%); Eastern Mediterranean 46.94% (95% CI 38.80% to 55.17%); South-East Asia 49.73% (95% CI 38.88% to 60.60%); Europe 39.82% (95% CI 27.26% to 53.09%); Western Pacific 39.21% (95% CI 25.78% to 53.52%); Africa 40.54% (95% CI 23.92% to 58.33%). - Time periods: 2016–2023: 45.33% (95% CI 39.74% to 50.98%); 1998–2015: 45.23% (95% CI 34.92% to 55.75%). - Specific micronutrients (random-effects, double arcsine): Vitamin D (k=66; n=27,169): 60.45% (95% CI 55% to 65%); Magnesium (k=16; n=3,210): 41.95% (95% CI 27% to 56%); Iron (k=3; n=1,887): 27.81% (95% CI 7% to 55%); Vitamin B12 (k=34; n=14,433): 22.01% (95% CI 16.5% to 27%). - Vitamin B12 among metformin users (k=30): 28.72% (95% CI 21.08% to 36.37%), higher than overall B12 deficiency. - Meta-regression: No significant overall temporal trend (slope ~0.0003, p=0.967). Non-significant slight increases over time for vitamin D (slope 0.0112, p=0.1027) and B12 (0.0065, p=0.4612); slight non-significant decrease for magnesium (-0.0098, p=0.5133). - Certainty of evidence (GRADE): Moderate, due to suspected publication bias despite otherwise not serious concerns in risk of bias, inconsistency, indirectness, and imprecision.

This review answers the research question by showing that nearly half of adults with T2D have at least one micronutrient deficiency, with vitamin D being the most prevalent. The higher prevalence among women and variations across WHO regions suggest influences of dietary patterns, lifestyle, cultural practices, and possibly health system factors. The elevated prevalence in hospital-based samples underscores potential selection bias and the burden seen in clinical populations. The findings reinforce biological plausibility that micronutrient status affects insulin action and glucose metabolism and may contribute to T2D progression and complications. However, causality cannot be inferred due to predominance of cross-sectional designs and limited control for confounding. Clinically, these results support consideration of assessing micronutrient status—particularly vitamin D, magnesium, and vitamin B12 (notably in metformin-treated patients)—as part of comprehensive T2D care, while acknowledging the need for rigorous trials to determine benefits of correction on glycaemic control and outcomes. For policy and research, the study highlights substantial global burden and regional differences, guiding prioritization of nutrition-focused interventions and surveillance.

The pooled global prevalence of any micronutrient deficiency among adults with T2D is 45.30%, with vitamin D deficiency (60.45%) and magnesium deficiency (41.95%) most common; vitamin B12 deficiency is frequent, particularly among metformin users (28.72%). Women have higher prevalence than men, and prevalence varies by WHO region. Given the predominance of hospital-based, cross-sectional studies, results should not be generalized to the broader population. Future research should prioritize well-designed population-based studies, prospective cohorts, and randomized trials to clarify causal relationships and evaluate the impact of screening and correction of deficiencies on T2D control, complications, and quality of life.

- Most included studies were hospital-based and cross-sectional, introducing selection bias and limiting causal inference and generalizability. - Few community-based studies prevented robust comparison with the general population. - Many studies inadequately described sampling frames, techniques, and sample size estimation; confounders were often not assessed. - High heterogeneity across studies; potential publication bias indicated by Egger’s test and asymmetrical funnel plot. - Variation in deficiency cut-offs and measurement methods across studies. - Despite comprehensive searching, some relevant studies may have been missed.