Prevalence and risk factors of nutritional anaemia among adolescents in India: a systematic review with meta-analysis of prevalence

Abstract

Background

Nutritional anaemia is a major public health concern among adolescents in India, threatening physical growth, cognitive development, and future maternal health. This systematic review with meta-analysis of prevalence aimed to estimate the pooled prevalence of nutritional anaemia among Indian adolescents aged 10–19 years and identify key associated risk factors.

Methods

A systematic search was conducted across PubMed, Embase, Web of Science, and Scopus up to February 2025. Data were extracted on study design, setting, diagnostic methods, prevalence, and risk factors. Quality was assessed with a modified Newcastle-Ottawa Scale. A random-effects meta-analysis estimated pooled prevalence, with subgroup analyses by state and gender, and pooled odds ratios for risk factors.

Results

Forty-five studies encompassing diverse Indian regions and 159,979 adolescents were included. The pooled prevalence of nutritional anaemia was 56% (95% CI: 49%–63%), with higher rates among girls (62%) than boys (39%). Iron deficiency (OR 2.38–4.68), other micronutrient deficiencies, low socioeconomic status, poor dietary diversity, female gender, and inadequate supplementation were consistently associated with higher anaemia risk.

Conclusion

Nutritional anaemia impacts more than half of Indian adolescents, with notable regional and gender differences. Its complex nutritional, socioeconomic, and behavioural causes demand targeted, context-specific interventions to enhance adolescent health nationwide.

KEY-POINTS

• There is a high prevalence of nutritional anaemia among Indian adolescents, especially in the states of Bihar and Uttar Pradesh, and Punjab.

• Around 56% of Indian adolescents have any one type of anaemia, most commonly iron-deficiency anaemia.

• Multi-micronutrient supplementation and food fortification should be tailored to local dietary patterns to reduce the prevalence of anaemia.

1. Background

Nutritional anaemia remains a significant global public health concern, with iron deficiency anaemia (IDA) being the most prevalent and widespread nutritional disorder. Adolescents are particularly vulnerable to nutritional anaemia due to increased physiological demands for iron during growth spurts, the onset of menstruation in females, and often inadequate dietary intake [1]. In low- and middle-income countries, including India, these vulnerabilities are compounded by poverty, gender inequality, food insecurity, and insufficient access to health care and nutrition education [2]. Despite numerous national health and nutrition programs, anaemia continues to affect a substantial proportion of Indian adolescents, thereby undermining their physical growth, cognitive development, and academic performance [3].

Most of the published studies on nutritional anaemia among adolescents included participants aged 10–19 years because the WHO defines adolescence within this age range. Although early and late adolescents differ physiologically, analysing the full adolescent spectrum allows researchers to capture variations in growth, puberty, dietary needs, and iron requirements, providing a broader understanding of nutritional anaemia prevalence and associated factors [1–4]. India is home to the largest adolescent population in the world, with approximately 253 million individuals aged 10 to 19 years. The National Family Health Survey (NFHS-5) estimates suggest that over 40% of adolescent girls and 20% of adolescent boys in India are anaemic, with higher prevalence noted in rural areas and among socioeconomically disadvantaged groups [4]. However, these estimates are based on cross-sectional surveys that often vary in methodology, sampling, and definition of anaemia. Furthermore, there is considerable heterogeneity in the prevalence of nutritional anaemia across different regions, age groups, and genders, indicating a need for a more granular understanding of the epidemiologic burden and its associated risk factors [5].

Nutritional anaemia in adolescence has both short- and long-term health consequences. In the short term, adolescents with anaemia may experience fatigue, reduced exercise tolerance, impaired cognitive function, and lower educational attainment [6]. In the long term, anaemic adolescent girls are at greater risk of adverse maternal and neonatal outcomes when they become pregnant, including preterm birth, low birth weight, and perinatal mortality. The economic implications are also significant, as anaemia can impair productivity and perpetuate cycles of poverty and malnutrition [7].

Efforts to address adolescent anaemia in India include iron and folic acid supplementation programs, food fortification initiatives, school-based health and nutrition interventions, and awareness campaigns [7]. The government of India has implemented the Weekly Iron and Folic Acid Supplementation (WIFS) program, the Anaemia Mukt Bharat (AMB) initiative, and components of the Integrated Child Development Services (ICDS) and Rashtriya Kishor Swasthya Karyakram (RKSK) [8]. Despite these efforts, the persistent high prevalence of anaemia among adolescents suggests that existing strategies may be insufficient, poorly implemented, or inadequately tailored to the local contexts in which adolescents live and learn.

This systematic review with meta-analysis of prevalence aims to estimate the pooled prevalence of nutritional anaemia among adolescents aged 10 to 19 years in India and identify and quantify the major risk factors associated with anaemia in this population. Due to heterogeneity in reporting of risk factors and effect measures, a quantitative synthesis (meta-analysis) of risk factors was not feasible; therefore, a narrative synthesis was conducted. By synthesising findings from studies conducted across diverse Indian settings, this review seeks to provide comprehensive insights that will support policymakers, clinicians, and public health practitioners in designing context-specific interventions to reduce the burden of adolescent anaemia and improve health outcomes.

2. Methods

This systematic review with meta-analysis of prevalence was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Supplementary Table 1) [9]. The protocol for this review was registered in the International Prospective Register of Systematic Reviews (PROSPERO) under the registration number CRD420251011650.

2.1. Eligibility criteria

Studies were eligible for inclusion if they met the following criteria: (i) original research articles reporting the prevalence of nutritional anaemia or its associated risk factors among adolescents (aged 10–19 years) in India; (ii) studies conducted in community or school settings, including cross-sectional, cohort, or case-control designs; (iii) studies employing standard definitions of anaemia based on haemoglobin concentration, as defined by the World Health Organization (WHO) or comparable criteria; (iv) studies published in English; and (v) studies with full-text availability. Studies were excluded if they (i) included participants outside the 10–19-year age range without disaggregated data for adolescents; (ii) focused exclusively on clinical populations with underlying diseases such as thalassemia, sickle cell anaemia, or chronic kidney disease; (iii) were reviews, editorials, commentaries, case reports, or conference abstracts without primary data; or (iv) lacked sufficient methodological details or haemoglobin assessment methods (v) studies with incomplete data.

2.2. Search strategy

A comprehensive literature search was conducted using three electronic databases: PubMed, Embase, Scopus, and Web of Science (Supplementary Table 2). The search strategy included both Medical Subject Headings (MeSH) and free-text terms related to “nutritional anaemia,” “iron deficiency,” “adolescents,” and “India.” Boolean operators (AND, OR) were used to combine keywords. The final search was conducted on February 4, 2025, with no date restrictions applied to capture all relevant studies to date. The search terms included combinations such as: (“anaemia” OR “iron deficiency anaemia” OR “nutritional anaemia”) AND (“adolescents” OR “teenagers” OR “youth”) AND (“India”). Additional studies were identified by hand-searching the reference lists of included articles and relevant reviews.

2.3. Screening and data extraction

All records retrieved from the database searches were imported into EndNote reference management software, and duplicates were removed. Two reviewers independently screened titles and abstracts for relevance. Full texts of potentially eligible studies were then reviewed in detail to determine final inclusion. Discrepancies were resolved through discussion or by consulting a third reviewer. Data extraction was conducted using a standardised, pilot-tested data extraction form. The following information was extracted from each study: author name, year of publication, study design, geographic location, setting (rural/urban), sample size, age range, gender distribution, method of haemoglobin estimation, anaemia classification criteria, prevalence rates, and reported risk factors with corresponding effect measures (odds ratios, relative risks, etc.).

2.4. Critical appraisal

The methodological quality and risk of bias of the included studies were assessed using a modified version of the Newcastle-Ottawa Scale (NOS) adapted for cross-sectional studies [10]. The scale evaluates three broad domains: (1) selection of the study population (representativeness and sample size), (2) comparability of groups (consideration of confounders), and (3) outcome assessment (measurement tools, definition of anaemia, and statistical analysis). Each study was awarded a maximum of 6 points. Studies were categorised as having low (8–10), moderate (5–7), or high (≤4) risk of bias. The methodological quality and risk of bias of the included studies were assessed using a modified Newcastle-Ottawa Scale (NOS) adapted for cross-sectional studies. The scale evaluates selection (representativeness and sample size), comparability, and outcome assessment (anaemia definition and measurement methods). Each study was assigned a maximum score of 6, and categorized as low (5–6), moderate (3–4), or high (≤2) risk of bias. A random-effects model (DerSimonian–Laird method) was used to account for between-study heterogeneity. Proportions were stabilised using the Freeman–Tukey double arcsine transformation. Statistical analyses were conducted using RevMan version 5.4 and MetaXL.

Sensitivity analyses were performed by excluding studies with small sample sizes, high risk of bias, and non-standard haemoglobin assessment methods; the pooled prevalence remained directionally consistent, although heterogeneity persisted. In addition, meta-regression analyses were conducted to explore methodological sources of heterogeneity, including diagnostic method, study setting (rural/urban/mixed), sampling strategy, and study population type. Two reviewers conducted the appraisal independently, and any disagreements were resolved by consensus or adjudication by a third reviewer (Supplementary Table 3).

3. Results

3.1. Literature search

A comprehensive literature search across PubMed, Embase, Web of Science, and Scopus databases initially identified 2,835 records. After removing 226 duplicates, 2,609 unique records remained for screening. Of these, 2,487 records were excluded based on title and abstract screening, leaving 122 reports for full-text review. No reports were unretrievable. Upon detailed assessment, 77 full-text articles were excluded for the following reasons: review articles (n = 21), editorials or opinion pieces (n = 16), studies lacking the population of interest (n = 18), or not reporting relevant outcomes (n = 22). Ultimately, 45 studies met the eligibility criteria and were included in the meta-analysis. This rigorous multi-step selection process is detailed in the PRISMA flow diagram (Figure 1).

Figure 1.

PRISMA flow diagram.

3.2. Summary of study characteristics

A total of 45 studies were included in this systematic review and meta-analysis, encompassing a diverse range of settings and regions across India (Supplementary Table 4). The majority of studies employed a cross-sectional design and were conducted in various environments, including schools, colleges, community settings, urban slums, rural villages, and both government and private institutions. The study populations primarily consisted of adolescents aged 10 to 19 years, with several studies focusing exclusively on girls, while others included both boys and girls, or reported data disaggregated by gender. Geographically, the studies spanned multiple states and regions, such as Maharashtra, Andhra Pradesh, Tamil Nadu, Himachal Pradesh, Uttar Pradesh, Gujarat, Haryana, Bihar, West Bengal, Rajasthan, Kerala, Karnataka, Punjab, Delhi, and all-India surveys. The sample sizes varied widely, ranging from small cohorts of fewer than 50 participants to large national datasets comprising more than 90,000 individuals. Most studies utilised the World Health Organisation (WHO) haemoglobin cutoffs for defining anaemia, with diagnostic methods including Sahli’s haemoglobinometer, cyanmet haemoglobin method, HemoCue analysers, automated haematology analysers, and immunoassays. According to the WHO anaemia guidelines, the haemoglobin cut-off for defining anaemia is based on the 5th percentile of haemoglobin values in a healthy reference population. Hence, the “acceptable degree of variability” is 5% from the cut-off value. Both rural and urban populations were represented, and a variety of sampling strategies were reported. Collectively, these studies provide a comprehensive overview of the prevalence and assessment of nutritional anaemia among Indian adolescents, highlighting significant regional, demographic, and methodological diversity. Studies with overlapping samples were screened non-statistically by selecting the most comprehensive, recent, and high-quality studies with manual data splitting, transparent reporting, and narrative synthesis.

3.3. Forest plot of pooled prevalence of nutritional anaemia among adolescents in India (random-effects model)

Figure 2 presents the forest plot of the prevalence of nutritional anaemia among adolescents in India across the 45 included studies. The pooled analysis, using a random-effects model, demonstrated a summary prevalence of 56% (95% CI: 49-63%). The random-effects (RE) model is significant in the pooled analysis of data, particularly in meta-analysis, primarily because it accounts for heterogeneity or variability between studies/groups and allows for the generalization of findings to a broader population. The prediction interval ranged widely from 12% to 95%, reflecting substantial heterogeneity across studies (I² = 99.5%, p < 0.001). Given the extremely high heterogeneity (I² > 99%), the pooled prevalence should be interpreted with caution. The wide prediction interval reflects substantial variability across study populations, methodologies, and regional contexts. Therefore, subgroup analyses were emphasized to provide more meaningful insights. Individual study prevalence estimates varied considerably, from as low as 10% to as high as 100%, with most studies reporting prevalence rates between 30% and 80%. The huge difference in the prevalence of anaemia is mainly due to the lifestyle, dietary pattern, geographical difference, food preference, and socioeconomic status of people living in different states in India, which play a major role in the development of nutritional anaemia among them. The weight of each study was similar, given the large number of events and participants, but substantial heterogeneity was observed (Tau² = 0.0587, Chi² = 9689.37, df = 44). To evaluate inter-study variability, values of I² greater than 50% or p < 0.1 were interpreted as indicating substantial heterogeneity. These results indicate that more than half of Indian adolescents included in the analysis were affected by nutritional anaemia, with prevalence estimates differing significantly by region, setting, and study methodology.

Figure 2.

Forest plot of the overall pooled prevalence of nutritional anaemia.

3.4. State-wise subgroup analysis

A state-wise subgroup analysis of the included studies revealed substantial geographic variability in the prevalence of nutritional anaemia among Indian adolescents (Figure 3). The pooled prevalence was highest in Bihar and Uttar Pradesh (53%; 95% CI, 0.53–0.54), followed by Punjab (73%; 95% CI, 0.01–1.00) and Maharashtra (53%; 95% CI, 0.37–0.69). States such as Tamil Nadu (61%; 95% CI, 0.41–0.89), Andhra Pradesh (68%; 95% CI, 0.41–0.89), and Gujarat (69%; 95% CI, 0.55–0.81) also reported high pooled prevalence estimates. In contrast, markedly lower prevalence rates were observed in Kerala (26%; 95% CI, 0.17–0.37) and West Bengal (35%; 95% CI, 0.28–0.42). National-level studies and those aggregating multi-state data indicated a pooled prevalence of 44% (95% CI, 0.29–0.59). Significant heterogeneity was evident within and between states (I² values >90% for most subgroups), reflecting diverse sociodemographic, dietary, and methodological factors. The test for subgroup differences was statistically significant (χ² = 1570.81, p < 0.0001), underscoring considerable regional disparities in the burden of nutritional anaemia among adolescents across India. Some of the states have wide confidence intervals (CIs) in subgroup analysis, signal low precision and high uncertainty on the prevalence of nutritional anaemia.

Figure 3.

Forest plot of state-wise subgroup analysis of nutritional anaemia.

3.5. Gender wise subgroup analysis

The pooled prevalence of nutritional anaemia among Indian adolescents reveals a striking gender disparity and a substantial overall public health burden (Figure 4a and 4b). Among adolescent girls, the pooled prevalence was 62% (95% CI, 54%–70%), with most studies reporting rates exceeding 50%. This analysis, which included 74,242 events among 139,150 participants, underscores that nearly two-thirds of adolescent girls are affected by nutritional anaemia. In contrast, among adolescent boys, the pooled prevalence was 39% (95% CI, 15%–66%) based on 3,851 cases out of 14,887 participants. While still significant, this figure is notably lower than that observed in girls. Individual study estimates for both genders showed wide variability, with prediction intervals ranging from 14% to 98% for girls and 0% to 100% for boys, and substantial heterogeneity was observed across all studies (I² > 99%). Out of 45 studies on nutritional anaemia, 38 studies were focusing on girls (vs 9 studies on boys). This disparity indicates that the researcher presumed that nutritional anaemia is more common among girls than boys, and many studies support this prevalence of nutritional anaemia in girls. These findings highlight that nutritional anaemia remains highly prevalent and persistent among Indian adolescents, disproportionately affecting girls, and suggest the influence of varied regional, socioeconomic, and methodological factors across studies.

Figure 4.

a. Forest plot of the prevalence of nutritional anaemia among boys.

Figure 4.b. Forest plot of the prevalence of nutritional anaemia among girls.

3.6. Associated risk factors

The analysis of risk factors associated with nutritional anaemia among Indian adolescents identified iron deficiency as the most robust and consistent predictor, with studies reporting significantly elevated odds (e.g. OR 2.38, 95% CI 1.29–4.40; OR 4.68, 95% CI not specified). Other micronutrient deficiencies—including vitamin B12 deficiency (OR 2.60, 95% CI 1.25–5.39), vitamin A deficiency (OR 1.86, 95% CI not specified), and zinc deficiency (OR 1.32, 95% CI not specified), were also independently associated with an increased risk of anaemia. Socioeconomic determinants played a critical role, with low socioeconomic status (e.g. low SES: OR 3.6, 95% CI 1.0–14.7; poor household status: OR 1.42, 95% CI 1.32–1.54), non-nuclear family structure (OR 1.7, 95% CI 0.98–3.2), parental illiteracy, and low household income emerging as significant contributors. Dietary factors further exacerbated anaemia risk, including low dietary diversity (OR 1.76, 95% CI 1.01–3.14), adherence to a vegetarian diet (OR 4.41, 95% CI 2.04–9.51), irregular iron–folic acid supplementation (aOR 1.39, 95% CI 1.08–1.82), and frequent consumption of tea or coffee with meals (aOR 1.41, 95% CI 1.21–1.66). Female gender (OR 2.179, 95% CI 1.551–3.061), increasing age (OR 1.41 per year, 95% CI 1.17–1.70), and late adolescence (e.g. boys with moderate/severe anaemia: RRR 3.68, 95% CI 2.45–5.53) were associated with greater odds of moderate or severe anaemia. Additional risk factors included underweight status (OR 1.74, 95% CI 0.87–3.48; low BMI: OR 1.78, 95% CI 1.68–1.89), rural residence (OR 0.49 for rural vs urban, 95% CI 0.34–0.70), worm infestation (OR 4.15, 95% CI 1.18–14.6), and lack of awareness or education. Protective associations were noted for regular fruit intake (≥3 times/week; OR 0.35, 95% CI 0.23–0.54). The considerable heterogeneity in risk factors and confidence intervals across studies highlights the multifactorial and context-dependent nature of nutritional anaemia in this population and underscores the need for comprehensive, multi-faceted intervention strategies (Figure 5).

Figure 5.

Summary Forest plot of pooled adjusted odds ratios (aOR) for harmonised risk factors of nutritional anaemia.

Legend: This plot summarizes the direction and magnitude of effect for evaluated risk factors using strictly adjusted odds ratios. Dark grey circles represent statistically significant predictors (p < 0.05), while light grey circles denote non-significant predictors. The shaded funnel background illustrates the prediction intervals for each pooled estimate.

3.7. Meta-regression analysis

Meta-regression analysis was performed (Table 1) to explore potential sources of heterogeneity across studies. The results indicated that studies conducted among female-only populations were significantly associated with a higher prevalence of anaemia (β = 0.092, p = 0.041). Similarly, studies conducted in rural settings demonstrated significantly higher prevalence estimates (β = 0.118, p = 0.028).

Table 1.

Meta-regression of heterogeneity.

Variable	Coefficient (β)	Standard Error (SE)	p-value
Sample size (continuous)	−0.00008	0.00004	0.061
Female-only population	0.092	0.045	0.041*
Rural setting	0.118	0.052	0.028*
HemoCue method	0.071	0.036	0.049*
Year of publication	−0.006	0.004	0.112

^*Statistically significant (p < 0.05).

Among the included studies (n = 45), the haemoglobin estimation methods depicted in Table 2, varied considerably. Approximately 13.3% used the cyanmethemoglobin method, 17.8% used HemoCue, 2.2% used Sahli’s method, and 13.3% used automated analysers or other techniques. Notably, 53.4% of studies did not clearly report the haemoglobin estimation method, which may have contributed to methodological heterogeneity in prevalence estimates.

Table 2.

Distribution of haemoglobin estimation methods across studies.

Haemoglobin estimation method	Number of studies (n)	Percentage (%)
Cyanmet haemoglobin method	06	13.3
HemoCue method	08	17.8
Sahli’s method	01	02.2
Automated analyser/others	06	13.3
Not reported / unclear	24	53.4
Total	45	100*

^*Percentages were calculated based on the total number of included studies (n = 45).

The use of HemoCue for haemoglobin estimation was also associated with a higher reported prevalence (β = 0.071, p = 0.049). In contrast, sample size and year of publication were not significantly associated with variability in prevalence estimates (p > 0.05). These findings suggest that demographic composition, geographic setting, and diagnostic methodology contribute to the substantial heterogeneity observed across studies.

3.8. Publication bias

Assessment of publication bias using a Doi plot with the Luis Furuya-Kanamori (LFK) index is presented in Figure 6. The LFK index value of 1.77 indicates minor to moderate asymmetry in the distribution of effect sizes, suggesting the potential presence of publication bias among the included studies. While values within ±1 generally reflect no asymmetry, an LFK index between ±1 and ±2 denotes minor to moderate asymmetry. This finding implies that although some publication bias may exist in the current evidence base for nutritional anaemia among Indian adolescents, the degree of asymmetry does not reach the threshold for major concern. Nevertheless, these results highlight the importance of cautious interpretation of pooled prevalence estimates, given the possible influence of unpublished or selectively reported data.

Figure 6.

Doi plot for assessment of publication bias.

4. Discussion

Recent systematic reviews and meta-analyses indicate that nutritional anaemia remains highly prevalent and persistent among adolescents in India, representing a significant public health concern across both rural and urban settings. The evidence consistently identifies iron deficiency as the primary cause, with additional contributions from deficiencies of folate and vitamin B12. A key limitation of this review is that only a limited number of included studies assessed micronutrient biomarkers such as serum ferritin, vitamin B12, and folate levels. As a result, the etiological attribution of anaemia to specific micronutrient deficiencies could not be consistently established, which may affect the interpretation of risk factor associations. Key risk factors include low dietary diversity, poor socioeconomic status, female sex (particularly post-menarche), inadequate sanitation, parasitic infections, and limited access to nutrition and health services [4,12]. These findings underscore the need for integrated, adolescent-focused nutrition interventions, including dietary diversification, micronutrient supplementation, and strengthening of public health programs.

Although some studies reported higher odds ratios for vitamin B12 deficiency, iron deficiency was more consistently assessed and is the most common global cause of anaemia. Its direct role in haemoglobin synthesis and higher prevalence in adolescents support its classification as the most robust predictor.

The present systematic review with meta-analysis of the prevalence of the 45 studies published earlier corroborates the persistent high prevalence of nutritional anaemia among Indian adolescents, with a pooled prevalence of 56% (95% CI, 49%–63%). This finding closely aligns with recent national survey data and underscores that nutritional anaemia remains a pervasive public health challenge affecting more than half of the adolescent population, especially among adolescents who are between 10 and 19 years of age, despite decades of targeted interventions. Notably, there is a striking gender disparity, with a higher burden among girls (62%, 95% CI, 54%–70%) compared to boys (39%, 95% CI, 15%–66%). This pattern parallels trends reported in the National Family Health Survey (NFHS), which noted increasing anaemia rates among adolescent girls (from 54% in NFHS-4 to 59% in NFHS-5) and boys (from 29% to 31%). Such disparities are likely explained by increased iron requirements in girls due to menstruation and pubertal onset, as well as prevailing social and cultural norms that prioritise male nutrition and health within households.

The meta-regression findings provide important insights into the sources of heterogeneity observed in this review. Higher prevalence estimates among female-only populations are consistent with known biological and nutritional vulnerabilities, including menstrual blood loss and increased iron requirements during adolescence. The significantly higher prevalence observed in rural settings likely reflects disparities in socioeconomic status, dietary diversity, access to healthcare, and implementation of public health interventions.

The substantial proportion of studies lacking clear reporting of haemoglobin estimation methods represents an important methodological limitation, potentially introducing measurement bias and contributing to the high heterogeneity observed across studies. Additionally, variation in haemoglobin estimation methods, particularly the use of HemoCue, may have contributed to differences in reported prevalence, highlighting the need for standardisation of diagnostic approaches across studies. Although sample size and year of publication were not statistically significant predictors, the overall findings underscore the multifactorial nature of heterogeneity in anaemia prevalence across the Indian adolescent population.

Previous meta-analyses and regional studies have reported pooled anaemia prevalence estimates among Indian adolescents ranging from 40% to 55%, with considerable heterogeneity based on geography, setting, and population characteristics (10). These earlier studies, along with national surveys, have consistently identified iron deficiency as the predominant risk factor, with additional contributions from deficiencies in other micronutrients such as folate, vitamin B12, and vitamin A, as well as low socioeconomic status, inadequate dietary diversity, early adolescence, female sex, menstrual blood loss, undernutrition, and parasitic infections [13]. School-based iron supplementation, food fortification, and public health campaigns have been implemented with varying degrees of effectiveness, yet national and regional data indicate only modest reductions in anaemia prevalence over recent decades [14].

In the current review, comprehensive subgroup analyses by state and gender provided a more granular understanding of the geographic and demographic heterogeneity in anaemia prevalence. Although Punjab showed a high pooled estimate, the wide confidence interval indicates low precision; therefore, more reliable estimates were observed in Bihar and Uttar Pradesh. Punjab showed (73%, 95% CI, 0.01–1.00) and in Bihar and Uttar Pradesh (53%, 95% CI, 0.53–0.54), with lower rates in Kerala (26%, 95% CI, 0.17–0.37) and West Bengal (35%, 95% CI, 0.28–0.42). These state-level differences likely reflect variations in dietary patterns, health service access, coverage, and quality of anaemia control programs, and underlying socioeconomic disparities. The extremely wide prediction intervals and substantial heterogeneity (I² > 99%) observed in this review highlight the contextual complexity and methodological diversity in the Indian literature on adolescent anaemia.

Analysis of risk factors revealed that iron deficiency was the most robust and consistent predictor, with pooled odds ratios ranging from 2.38 (95% CI, 1.29–4.4) to 4.68. Other micronutrient deficiencies, including vitamin B12 (OR 2.60, 95% CI, 1.25–5.39), vitamin A (OR 1.86), and zinc (OR 1.32), were also significantly associated with greater anaemia risk, supporting the need for integrated multi-micronutrient approaches in prevention and treatment. Socioeconomic factors remained critical determinants, with low socioeconomic status (OR 3.6, 95% CI, 1.0–14.7), poor household income (OR 1.42, 95% CI, 1.32–1.54), non-nuclear family structure (OR 1.7, 95% CI, 0.98–3.2), and rural residence (OR 0.49, 95% CI, 0.34–0.70) all increasing risk. Dietary risk factors included low dietary diversity (OR 1.76, 95% CI, 1.01–3.14), vegetarian diet (OR 4.41, 95% CI, 2.04–9.51), irregular iron-folic acid supplementation (aOR 1.39, 95% CI, 1.08–1.82), and frequent tea or coffee consumption with meals (aOR 1.41, 95% CI, 1.21–1.66). Protective associations were observed for regular fruit intake (OR 0.35, 95% CI, 0.23–0.54). The multifactorial and overlapping nature of these risk factors underscores the importance of both biomedical and social interventions in addressing nutritional anaemia in this population. After 50 years of many different government programs, rates of anaemia, especially iron deficiency anaemia, have not diminished, and are in fact rising. This calls for a more aggressive approach among clinicians and educators, with special emphasis on new technology and addressing supply chain issues.

The high prevalence of nutritional anaemia among Indian adolescents, particularly girls, highlights the urgent need for strengthened, context-sensitive intervention strategies. While national programs such as Anaemia Mukt Bharat, Weekly Iron and Folic Acid Supplementation, and food fortification have made progress, their impact has been limited by inconsistent implementation, coverage gaps, and suboptimal compliance. Adolescent-specific and gender-responsive interventions are essential, including menstrual health management, dietary counselling, and accessible iron and folate supplementation. School-based health platforms present a valuable opportunity for large-scale intervention, but must be supported by effective monitoring, supply chain management, and robust community engagement to maximise impact. The study recommends that policymakers take necessary steps to reduce the incidence and prevalence of nutritional anaemia among adolescents, especially girls and people living in poor socioeconomic backgrounds, rural areas, and underdeveloped states, to minimise the future risks associated with nutritional anaemia in the country.

Efforts to address the social determinants of anaemia, including poverty reduction, parental education (especially for mothers), and empowering adolescents with nutrition and health literacy, are also critical. Behavioural change communication and community mobilisation, potentially leveraging digital health tools, may enhance awareness and uptake of preventive services. In regions with multiple overlapping deficiencies, multi-micronutrient supplementation and food fortification should be tailored to local dietary patterns and resource availability. Integration of deworming, dietary diversification, and routine monitoring of haemoglobin and micronutrient status should be embedded in adolescent health services.

State-level heterogeneity in anaemia prevalence further highlights the need for decentralised planning and implementation, with high-burden states prioritised for additional investment, operational research, and programmatic support. Coordination between national and state health authorities, along with engagement of non-governmental organisations and community stakeholders, will be essential for sustained progress.

This review has several strengths, including its comprehensive coverage of published studies from diverse regions, use of robust systematic review methodology, and assessment of both prevalence and risk factors with quantitative synthesis where possible. However, limitations must be acknowledged. Substantial heterogeneity and wide prediction intervals suggest caution in generalising pooled estimates. Most included studies were cross-sectional, limiting causal inference, and there was inconsistency in the reporting and adjustment of risk estimates. The presence of minor to moderate publication bias, as indicated by the LFK index, further suggests that the burden of anaemia may be subject to selective reporting or publication of positive findings.

Well-designed longitudinal studies are warranted to elucidate causal pathways, establish temporal relationships, and evaluate the effectiveness of interventions in modifying the progression and determinants of anaemia among adolescents. Standardisation in diagnostic criteria, haemoglobin assessment, and risk factor reporting will improve comparability across studies. Qualitative and mixed-methods research is needed to explore cultural, behavioural, and health system barriers to anaemia prevention and control. Detailed, state-specific, and district-level analyses will further aid in developing tailored interventions. Innovative delivery models, including digital health interventions, community-based screening, and adolescent peer support, warrant exploration, and ongoing monitoring and operational research should be embedded within national and state anaemia control programs.

A key limitation of this review is that only a limited number of included studies assessed micronutrient biomarkers such as serum ferritin, vitamin B12, and folate levels. As a result, the etiological attribution of anaemia to specific micronutrient deficiencies could not be consistently established, which may affect the interpretation of risk factor associations. Limited evidence was available regarding folate deficiency, and its association with anaemia was not consistently reported across the included studies, which may have limited a comprehensive assessment of its role in anaemia etiology.

5. Conclusions

The nutritional anaemia remains a critical public health challenge among Indian adolescents, with a pooled prevalence of 56% and pronounced disparities by gender and region. Iron deficiency and other micronutrient deficiencies, socioeconomic and dietary factors, and behavioural practices all contribute to this multifaceted problem. Meta-regression analysis further highlighted that gender, rural setting, and diagnostic methods significantly contribute to variability in anaemia prevalence, emphasizing the need for standardized methodologies and context-specific interventions. Strengthening the coverage, quality, and equity of adolescent anaemia control programs while addressing broader social determinants will be essential to achieving meaningful reductions in anaemia prevalence and improving the health and well-being of India’s adolescents.

Funding Statement

No funds, grants, or other support was received.

Consent for publication

Not Applicable

Ethical approval and consent to participate

Not Applicable

Disclosure statement

The authors have no competing interests to declare that are relevant to the content of this article.

Data availability statement

All data supporting the findings of this study are available within the paper and its Supplementary Information.