https://orcid.org/0000-0001-7110-3347
Figures
Abstract
Mortality following severe acute malnutrition (SAM) remains high in Africa. Some children with SAM have altered immune responses and impaired neurodevelopment; however, whether vitamin D deficiency contributes to these poor outcomes remains unclear. We conducted a cross-sectional analysis of circulating 25-hydroxyvitamin D [25(OH)D] concentrations at hospital discharge (25(OH)D) in 442 children aged 0–59 months with SAM at one hospital in Lusaka, Zambia, and two hospitals in Harare, Zimbabwe. Also, we investigated the relationship between 25(OH)D concentrations and host factors. In 442 children, plasma 25(OH)D concentrations were measured using ELISA. Vitamin D deficiency was identified in 33/442 (7.5%; 95% CI: 5.2–10.3) children, all of whom were from Zimbabwe. A multivariable regression model that included age, sex, HIV status, season, oedema, cerebral palsy, and country demonstrated significantly higher 25(OH)D levels among children aged 6–11 months (adjusted (adj) ratio; 1.3; 95% CI: 1.1, 1.6; P = 0.003), 12–23 months (adj ratio; 1.5; 95% CI: 1.3, 1.8; P < 0.001), and 24–59 months (adj ratio; 1.5; 95% CI: 1.3, 1.8; P < 0.001) compared with those <6 months. Sampling during the cool season (adj ratio; 1.2; 95% CI: 1.1, 1.3; P = 0.002), absence of cerebral palsy (adj ratio; 1.2; 95% CI: 1.1, 1.4; P = 0.002) and residence in Zambia (adj ratio; 1.2; 95% CI: 1.1, 1.2; P < 0.001) were associated with higher concentrations of 25(OH)D, even after excluding children <6 months. These findings indicate that vitamin D status among children with SAM is significantly influenced by age, seasonality, cerebral palsy, and geographical location. To facilitate management and improve clinical outcomes in these children, we recommend further investigation into whether the current vitamin D content in ready-to-use therapeutic foods is sufficient for all children and whether additional supplementation improves clinical outcomes, especially in high-risk groups. If deficiency is present, we recommend that children be managed in accordance with local guidelines and policies.
Citation: Phiri TN, Dumbura C, Mutasa K, Ngosa D, Chuulu N, Majo FD, et al. (2026) A cross‑sectional analysis of vitamin D status and associated factors in children recovering from severe acute malnutrition in Zambia and Zimbabwe. PLoS One 21(8): e0355404. https://doi.org/10.1371/journal.pone.0355404
Editor: Mohammed Hasen Badeso, PLOS: Public Library of Science, ETHIOPIA
Received: December 20, 2025; Accepted: July 21, 2026; Published: August 28, 2026
Copyright: © 2026 Phiri et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The complete de-identified dataset supporting the findings of this study is available in S8 Table in the Supplementary Material.
Funding: This study was supported by the Medical Research Council (UK) (grant MR/K012711/1), Wellcome Trust (grants 107634/Z/15/Z to MBD and 108065/Z/15/Z to AJP) and jointly by the Wellcome Trust and Royal Society (grant 206225/Z/17/Z to CDB). The funders had no role in the study design, data collection, data analysis, interpretation of the data, the decision to publish, or the preparation of the manuscript.
Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests. ARM declares receipt of funding in the last 36 months to support vitamin D research from the following companies who manufacture or sell vitamin D supplements: Pharma Nord Ltd, DSM Nutritional Products Ltd, Thornton & Ross Ltd and Hyphens Pharma Ltd. ARM also declares receipt of vitamin D capsules for clinical trial use from Pharma Nord Ltd, Synergy Biologics Ltd and Cytoplan Ltd; support for attending meetings from Pharma Nord Ltd and Abiogen Pharma Ltd; receipt of consultancy fees from DSM Nutritional Products Ltd and Qiagen Ltd; receipt of a speaker fee from the Linus Pauling Institute; participation on Data and Safety Monitoring Boards for the VITALITY trial (Vitamin D for Adolescents with HIV to reduce musculoskeletal morbidity and immunopathology, Pan African Clinical Trials Registry ref PACTR20200989766029) and the Trial of Vitamin D and Zinc Supplementation for Improving Treatment Outcomes Among COVID-19 Patients in India (ClinicalTrials.gov ref NCT04641195); and unpaid work as a Programme Committee member for the Vitamin D Workshop. All other authors declare no conflicts of interest.
Introduction
Severe Acute Malnutrition (SAM) encompasses two forms of acute malnutrition: severe wasting and nutritional oedema [1]. According to World Health Organisation (WHO) growth standards, SAM in children aged 6–59 months is defined as a mid-upper arm circumference (MUAC) below 115 mm or a weight-for-height z-score (WHZ) below −3 [2], with or without bilateral oedema. Children with SAM who have severe oedema, poor appetite, and/or medical complications such as diarrhoea, pneumonia or metabolic problems need urgent medical intervention and hospitalisation for management [1]. According to the WHO, an estimated 19 million children below the age of 5 years are living with SAM, and 400,000 die each year [3]. A review by Asebe et al. reported that 2.1% (95% CI: 2.0, 2.2%) of children aged between 6 and 59 months in sub-Saharan Africa are affected by severe wasting [4], while in Zambia, the Demographic and Health Survey (DHS) of 2024 reported that approximately 1% of Zambian children are living with SAM [5]. In Zimbabwe, approximately 15,000 children are treated for severe wasting annually [6].
SAM is accompanied by micronutrient deficiencies, including zinc, selenium, iron [7–9] and Vitamin D [10]. In children with SAM, Vitamin D helps maintain calcium and phosphorus balance by promoting their absorption in the intestines and kidneys [11]; while deficiency complicates recovery [10]. Children with SAM have an altered immune response to bacterial antigens [12], which increases their risk of mortality. Given that Vitamin D exerts anti-inflammatory and antimicrobial effects [13–17], understanding vitamin D status in children recovering from SAM may be clinically relevant in this high-risk group, as it may contribute to recovery. Besides clinical management, treatment of SAM includes nutritional supplementation via Ready-to-use therapeutic food (RUTF), which contains 15–22ug/100g of Vitamin D (Cholecalciferol) [18]. Whether this is sufficient for all children with SAM is unclear. In low- and middle-income countries (LMICs), responsiveness to RUTF may be impeded by environmental enteropathy, a condition characterised by increased malabsorption [19] and common among people living in unsanitary environments [20–22].
Vitamin D status is evaluated using serum 25‑hydroxyvitamin D (25(OH)D), the most abundant and stable circulating metabolite, with a half-life of approximately three weeks [23,24]. Using a cut-off of <20ng/mL (where nmol/L = ng/mL*2.5), a study in Tunisia found that 84 (97%) of the women and 85 (98%) of the neonates were vitamin D deficient, with 76 (87%) and 78 (90%) being severely deficient (<12ng/mL) [25]. Although the prevalence of vitamin D deficiency (<20 ng/mL) declined from 76.4% at 6 weeks of age, infants who were exclusively breastfed at 6 weeks of age had a higher risk of vitamin D deficiency than formula-fed infants [26]. In Nigerian mothers and their infants, the duration of exclusive breastfeeding was inversely associated with plasma vitamin D levels [27]. These findings not only provide evidence of high levels of vitamin D deficiency in some parts of Africa but also highlight a higher risk of vitamin D deficiency in exclusively breastfed infants as a high-risk group.
Circulating 25(OH)D levels vary with sunlight exposure and can exhibit seasonal patterns [28–30]. In Africa, these variations can be attributed to heterogeneous climates, diets, latitudes, geographies, and skin pigmentation [31]. Vitamin D deficiency and thresholds depend on several factors, including the population of interest and condition [32–34], which complicates comparisons across settings. However, evidence of vitamin D supplementation or deficiency in Zambian or Zimbabwean children is limited.
Although children with SAM are discharged from hospital only after clinical recovery and stabilisation [3,8], post-discharge mortality persists [35], potentially reflecting delayed immune restoration relative to nutritional recovery [12]. Children with SAM may also experience impaired neurodevelopment [36]; however, despite the known roles of vitamin D in immune function and neurodevelopment, vitamin D status at discharge remains poorly described. We aimed to i) determine the prevalence of 25(OH)D deficiency at discharge and ii) identify determinants associated with vitamin D concentrations in Zambian and Zimbabwean children recovering from SAM, thereby addressing an important gap in post-discharge care.
Materials and methods
Study design
This was a cross-sectional study nested in the Health Outcomes, Pathogenesis and Epidemiology of Severe Acute Malnutrition (HOPE-SAM) study, which has been described in detail elsewhere (https://osf.io/29uaw/.) [37,38]. Briefly, HOPE-SAM was a longitudinal prospective cohort study of 745 children under five years of age hospitalised for complicated SAM at the University Teaching Hospital in Zambia and Parirenyatwa General and Harare Children’s Hospitals in Zimbabwe, recruited between 14th July 2016 and 15th March 2018 [35].
The HOPE-SAM study enrolled 755 children with SAM across both countries. Of these, three were ineligible, two were co-enrolled in other studies, three exited before baseline and two died, leaving 745 children (242 in Zambia and 503 in Zimbabwe) who were included in the main study [35,37]. However, of the 745 children, 70 died, and 26 exited from the study while in hospital, leaving 649 children from both countries who were discharged alive (S1 Fig). Of the 649 children at discharge, only 442 had plasma samples available for this cross-sectional analysis of 25(OH)D in both Zambia (n = 101) and Zimbabwe (n = 341) (S1 Fig).
Inclusion and exclusion criteria
This sub-study included children at hospital discharge who had recovered clinically and were to be managed as outpatients after meeting the World Health Organisation (WHO) criteria of WHZ > –2 and/or MUAC >125mm without oedema for at least two weeks [39]. As screening was conducted in the main study, children in this cross-sectional analysis were excluded only if they had insufficient plasma samples for the 25(OH)D analysis.
Laboratory analysis
Plasma concentrations of 25(OH)D were quantified in singlicate using ELISA kits from the Immunodiagnostic System (reference number AC-575F1; Lot number J45789) according to the manufacturer’s instructions. Two assay controls provided with every kit were included and analysed with each plate, yielding comparable results between the two sites. Each plate was analysed using the same settings on Biotek ELx808 readers located at Tropical Gastroenterology and Nutrition Group (TROPGAN) laboratories and the Zvitambo Institute for Maternal and Child Health Research laboratory in Zambia and Zimbabwe, respectively. Sample concentrations were determined using a non-linear 4-parametric logistic regression (4-PL) standard curve.
Statistical analysis
Participant clinical and demographic characteristics were collected using a questionnaire that was completed by the caregiver and administered by the study nurse at baseline. Data were presented as medians with interquartile ranges (IQRs) or as proportions with 95% confidence intervals (CIs). For proportions, statistical significance was assessed using Pearson’s Chi-square test. As 25(OH)D concentrations were not normally distributed, the Mann-Whitney or Kruskal-Wallis tests were used to compare medians between and across groups, respectively, with P < 0.05 considered significant. Values of 25(OH)D ≥ 50 nmol/L were classified as sufficient [40,41], while concentrations < 50 nmol/L were considered deficient. Seasonality at discharge was assessed by categorising months into three seasons. In Zimbabwe [42], the rainy season is from November to April, the cool season from May to August, and the hot season from September to October. In Zambia [43], the rainy season is from December to April, the cool season from May to August, and the hot season from September to November. Seasons in both countries extensively overlap, with slightly longer cool and hot seasons in Zimbabwe and Zambia, respectively.
Despite the Shapiro-Wilk test indicating that the data were not normally distributed, a visual inspection of the 25(OH)D concentrations using a histogram and Q–Q plot revealed no substantial deviations from normality. Log transformation improved the distribution of the residuals, with Q–Q plots showing approximate normality, with minor deviations remaining at the tails. Variance inflation factors (VIFs) were examined to assess multicollinearity, and all VIFs were below 3 (S1 Table). Multivariable regression models were specified a priori based on known determinants (country, HIV infection, age, sex, oedema status, cerebral palsy and seasonality), with log-transformed 25(OH)D concentrations as the dependent variable. As children <6 months were more likely to have been exclusively breastfed, a sensitivity analysis excluding children <6 months was undertaken. All beta coefficients and 95% CIs are reported as adjusted (adj) ratios. All statistical analyses were performed using Stata 17.0 (Stata Corp, College Station, TX), and all graphs were created using GraphPad Prism 10.
Ethical considerations
Ethical approval for the study was obtained from the University of Zambia Biomedical Research Ethics Committee (010-02-16) and the Medical Research Council of Zimbabwe. The ethics committee of Queen Mary University of London, U.K., also provided a non-binding advisory review. Before any child could be enrolled in this study, written informed consent was obtained from the caregivers following careful explanation by trained study nurses using standardised consent forms reviewed and approved by our ethics committees. Comprehension of the consent form was checked using a checklist before consent forms were signed.
Results
Characteristics of the study participants
Among 442 children recruited in Zambia and Zimbabwe, we found that age, sex and clinical phenotype did not differ significantly (Table 1). Zambia had significantly more children who were living with HIV than Zimbabwe (27 (26.7%) and 52 (15.2%), respectively; P = 0.008).
25(OH)D concentrations in Zambia and Zimbabwe
Concentrations of 25(OH)D were higher in the Zambian (89.1 nmol/L; IQR 72.3, 107.7) than in Zimbabwean children (77.1 nmol/L; IQR 63.1, 95.9) by 12.0 nmol/L (95% CI 5.1, 18.9; P = 0.001; Fig 1). Between the two Zimbabwean hospitals, Parirenyatwa (78.9 nmol/L; IQR 62.3, 97.6) and Harare Children’s (75.8 nmol/L; IQR 63.7, 92.3) hospitals, plasma concentrations of 25(OH)D did not differ significantly. However, 25(OH)D concentrations from these hospitals were significantly lower than those from UTH (89.1 nmol/L; IQR 75.3, 107.7) (S2 Fig). Of the 442 children, we found concentrations consistent with deficiency in 33 (7.5%; 95% CI 5.2, 10.3), all of whom were from Zimbabwe. Hence, excluding all Zambian children, 33/341 (9.7%; 95% CI 6.8–13.3) of Zimbabwean children were deficient.
The Mann-Whitney test was used; Zambian (n = 101) and Zimbabwean (n = 341) children were included.
Effect of seasonality on circulating 25(OH)D concentrations
In Zambia and Zimbabwe combined, plasma 25(OH)D concentrations were significantly lower (P = 0.004) in the hot (69.3 nmol/L; IQR: 57.56–96.17) season than in the rainy (79.1 nmol/L; IQR: 65.4–98.0) or cool (88.4 nmol/L; IQR: 69.9–103.0) seasons (shown in Fig 2a). This seasonal effect was country-specific: Zimbabwe had significantly higher concentrations in the Cool season (84.2 nmol/L, IQR: 69.5–98.5; versus Hot season: 67.5 nmol/L, IQR: 57.3–88.0; versus Rainy season; 77.8 nmol/L; IQR: 63.3–96.0; P = 0.010) but the concentrations did not differ significantly among Zambian children (Cool: 103.0 nmol/L; IQR: 92.8–107.7; versus Hot: 88.4 nmol/L; IQR: 77.5–111.2; versus Rainy; 86.4 nmol/L; IQR: 71.0–106.7; P = 0.117) (shown in Fig 2b).
a. A combined comparison in both Zambian and Zimbabwean children according to the season when the samples were collected: cool (n = 73), hot (n = 88) and rainy (n = 280): Kruskal-Wallis was used. b. A separate comparison of 25(OH)D concentrations according to the season of sample collection in each country. Cool; Zambia (n = 13) vs Zimbabwe (n = 60); Hot; Zambia (n = 24) vs Zimbabwe (n = 64); Rainy; Zambia (n = 64) vs Zimbabwe (n = 216). Kruskal-Wallis was used.
Effect of clinical factors on circulating 25(OH)D concentrations
We found no evidence of a difference in 25(OH)D concentrations between male (81.2, IQR: 63.7–96.9) and female children (78.8, IQR: 65.3–98.5; 0.945: Fig 3e). However, 25(OH)D concentrations were significantly lower in children aged 1–5 months (49.1 nmol/L, IQR 43.5–70.7), 6–11 months (70.5 nmol/L, IQR 56.3–89.3 nmol/L) and 24–59 months (73.3 nmol/L, IQR 63.9–9.6 nmol/L) relative to those aged 12–23 months (79.8 nmol/L, IQR 66.2–96.8 nmol/L; P = 0.04; Fig 3d). 25(OH)D concentrations were higher in children with HIV (85.9 nmol/L, IQR: 69.6–104.4) than in children without HIV (78.6 nmol/L, IQR: 64.2–97.3; P = 0.05; Fig 3a). Children with cerebral palsy had lower (65.3 nmol/L, IQR 56.3–78.3) circulating 25(OH)D concentrations than children who did not have cerebral palsy (81.2 nmol/L, IQR 65.7–98.8; P < 0.001; Fig 3c). Circulating plasma concentrations of 25(OH)D did not differ between children admitted with oedematous (81.9 nmol/L; IQR: 68.4–105.3) versus non-oedematous SAM (79.1 nmol/L; IQR: 64.6, 97.0; P = 0.17; Fig 3b).
HIV+ (n = 79) and HIV- (n = 363); b). Oedema (n = 36) and no-Oedema (n = 405); c) Cerebral palsy (n = 27) and No Cerebral palsy (n = 415); d). Age: 1-5 months (n = 13), 6-11months (n = 63), 12-23months (n = 262), and 24-59 months (n = 103); Male (n = 234) and Female (n = 208).
Mortality
To determine if there was an association between vitamin D and mortality, plasma 25(OH)D concentrations at discharge were compared between the 36 children who were known to have died over the subsequent period of follow-up and the 406 children who were known to have survived. Plasma 25(OH)D concentrations were similar (P = 0.905) in those who survived (79.4 nmol/L, IQR: 64.9–98.0) and those who died (80.95 nmol/L, IQR: 62.6–102.8).
Multivariable linear regression analysis
To assess the relative importance of country, HIV infection, oedema status, age, sex, cerebral palsy, and seasonality, these variables were included in a regression model, with log-transformed 25(OH)D concentrations as the dependent variable. In this model, residence in Zambia (adj ratio; 1.2, 95%CI 1.1, 1.2; P < 0.001), being older than 6 months (adj ratio; 6–11 months: 1.3, 95% CI 1.1, 1.6; P = 0.003, 12–23 months: 1.5, 95% CI 1.3, 1.8; P < 0.001 and 24–59 months: 1.5, 95% CI 1.3, 1.8; P < 0.001), absence of cerebral palsy (adj ratio; 1.2, 95%CI 1.1, 1.4; P = 0.002) and sampling during the cool season (adj ratio; 1.2; 95% CI 1.1, 1.3; P = 0.002) were associated with significantly higher 25(OH)D concentrations (S2 Table). Using the same model, excluding children <6 months, residence in Zambia (adj ratio; 1.2, 95% CI 1.0, 1.2; P < 0.001), older than 11 months (adj ratio; 12–23months: 1.1, 95% CI 1.0, 1.2; P = 0.004 and 24–59 months: 1.1, 95% CI 1.0, 1.2; P = 0.015), absence of cerebral palsy (adj ratio; 1.2, 95% CI 1.1, 1.4; P = 0.002) and sampling during the cool season (adj ratio; 1.2; 95% CI 1.1, 1.3; P = 0.002) were associated with higher plasma concentrations of 25(OH)D (S3 Table).
Discussion
Vitamin D has anti-inflammatory and immunomodulatory properties [44] and, therefore, may have implications for clinical recovery. We set out to determine if deficiency is a feature of SAM in two countries in southern Africa. Among 442 children drawn from the HOPE-SAM study, at hospital discharge, 7.5% had vitamin D deficiency (<50 nmol/L), but none were from Zambia. Sampling during the cool season and older age were associated with higher circulating 25(OH)D concentrations, whilst cerebral palsy was associated with lower concentrations.
The prevalence of vitamin D deficiency we report here is lower than that reported in previous studies of malnourished and well-nourished school-going children elsewhere in Africa [25]. A study conducted on 117 malnourished and 41 non-malnourished children aged 6–24 months in Uganda reported comparable vitamin D levels between the two groups (32.5 nmol/L (±12.0 SD) and 32.2 nmol/L (10.9 SD); P = 0.868) [45]. In 828 hospitalised malnourished adults in Switzerland, the prevalence of vitamin D deficiency (<50 nmol/L) was 58.2%, and these patients were more likely to die before 180 days (odds ratio (OR) 1.42; 95% CI 1.03–1.94, P = 0.03) than adults with sufficient vitamin D levels [46]. These studies highlight the importance of Vitamin D in health and survival, suggesting that supplementation may enhance clinical outcomes.
In two separate studies involving older Mongolian school-going children, increases in Vitamin D levels after supplementation were observed, but with no significant differences between the Vitamin D-supplemented and placebo groups in terms of improvement in growth, body composition, or pubertal development [47], nor in height-for-age z-score or body mass index-for-age z-score in South African school-going children aged 6–11 years [48]. Nonetheless, vitamin D supplementation was associated with improved wasting, evidenced by significant weight gain in children with SAM [49]. Also, a clinical trial from Pakistan suggests that adjunctive vitamin D supplementation may improve outcomes in children receiving standard therapy for uncomplicated SAM and may be essential for growth recovery and neurodevelopment [50].
Currently, the prevalence of cerebral palsy is estimated to be around 3.4% in LMIC [51]. Hence, children who have both SAM and cerebral palsy, as observed in this and other studies [45,52–57], often have lower circulating 25(OH)D concentrations and may benefit from vitamin D supplementation. However, both conditions are complex, and clinical presentation and requirements may vary from one child to another. Children with cerebral palsy may require higher levels of vitamin D supplementation than those presenting with SAM only, as they already have pre-existing reduced bone mineral density, bone fragility, osteopenia, and rickets [56]. Hence, further investigations to determine optimal supplementation doses are warranted.
Possible reasons for the lower 25(OH)D concentrations in Zimbabwean children compared with Zambian children may include lower vitamin D intake, greater malabsorption, and/or inadequate sunlight exposure, as well as variations in cultural and socio-economic factors [58]. These factors, in turn, may include diet [59], dress, health-seeking and other behaviours. Additionally, all children <6 months old (n = 13) were from Zimbabwe, which may in part explain the differences. Although caregivers influence the feeding behaviour of children [60] and maternal practices being linked to infants’ health outcomes [61], in this study, primary caregiver type differed modestly between countries, with no evidence that caregiver type explained vitamin D deficiency; with only one child not cared for by a biological parent being vitamin D deficient. Lower vitamin D levels have previously been associated with colder seasons due to reduced sunlight exposure [62]. However, we report lower 25(OH)D concentrations during the hot season than the cool and rainy seasons, which may be explained by several factors, including behavioural avoidance of sun exposure, which may reduce cutaneous vitamin D synthesis despite higher ambient UV availability [63,64]. Although the differences may be minimal, Lusaka lies at 15.4°S and Harare at 17.8°S [65,66]. Furthermore, since 1930, annual temperatures in Harare have been 1–4°C cooler than in Lusaka, with higher precipitation in Zimbabwe than in Zambia (S1–S4 Tables) [67]. This may be explained by the altitude differences between Harare (1,483 m) and Lusaka (1,280 m). Though diets in Zambia and Zimbabwe are very similar [68], as are therapeutic formulations for treatment of SAM, it is also possible that the vitamin D content of agricultural products in the two countries vary seasonally [69], and the differences between countries could be multifactorial.
In this study, the youngest group (1–5 months) had the lowest circulating 25(OH)D concentrations, which may be attributed to exclusive breastfeeding and limited sunlight exposure due to age. A study in Tunisia among women and their neonates reported mean 25(OH)D serum levels of 6.82 ± 5.14 ng/mL (range 3.60–23.77) and 5.92 ± 4.15 ng/mL (range 3.60–22.28) (P = 0.001), respectively [25]. In Tanzanian children born to HIV-negative women, children <6 months who were exclusively breastfed had double the risk of Vitamin D deficiency compared to children who received formula, and younger children were more likely to have lower 25(OH)D concentrations [26]. Hence, re-evaluating whether the current levels of vitamin D in RUTF are sufficient for all children is essential, as younger children may need more than they receive currently, in which case RUTF reformulation or supplementation could be considered.
In this study, 25(OH)D plasma concentrations did not differ significantly between children who were living with HIV and children without HIV. However, 51/78 of the children living with HIV were taking antiretroviral therapy (ART). In contrast to our findings, previous studies have reported a higher prevalence of vitamin D deficiency among people living with HIV, with consequences including reduced bone mineral density, osteoporosis, and osteopenia [70–73]. We found no evidence of sex-related differences in vitamin D status; current evidence is inconsistent, with some studies reporting higher vitamin D levels in males than in females, and others reporting higher levels in females than in males [74–76]. Additionally, although we found no significant difference in 25(OH)D concentrations between oedematous and non-oedematous SAM, oedematous SAM is associated with poorer inpatient clinical outcomes [77]. A previous study reported that children with marasmus (non-oedematous SAM) were 10.8 times (OR 10.8, CI = 1.356–86.236, p = 0.03) more likely to have Vitamin D deficiency than those with kwashiorkor or marasmic-kwashiorkor [78], but our study did not find such a difference. Lastly, although we found no significant differences in 25(OH)D levels between children who survived or died during recovery, low plasma concentrations of 25(OH)D are inversely associated with all-cause mortality [79]. Although different from SAM, a study in children admitted to the paediatric intensive care unit reported higher vitamin D levels in survivors than in non-survivors [80]. Also, in children who were on dialysis, 64.7% of those who were vitamin D deficient died, whereas all those with sufficient levels survived [81]. Further investigations on the associations between HIV, sex, oedema status, mortality and vitamin D in malnourished children are needed.
There are significant limitations to this study. We did not have the opportunity to include a well-nourished community or hospital control group (admitted for reasons other than SAM). We also acknowledge that differences in sample sizes between Zambia and Zimbabwe reduce the power to draw firm conclusions across countries. This is especially true of cerebral palsy and oedema, as the numbers of affected children were small. Further investigation is needed to determine whether these subgroup differences persist. Lastly, although all experiments included assay controls and were analysed using the same ELISA reader with the same instrument settings, it is possible that running our samples singly and processing samples in two different laboratories may contribute to the difference in concentrations between Zambia and Zimbabwe.
Conclusion and recommendations
Vitamin D deficiency (<50 nmol/L) was present in 7.5% of children recovering from SAM at hospital discharge. We found no evidence that 25(OH)D concentrations differed by sex, HIV status, oedema status, or subsequent survival. In contrast, 25(OH)D concentrations were associated with age, country, season, and cerebral palsy. These findings suggest that vitamin D status at discharge varies across subgroups and may warrant targeted attention during recovery. Future studies should evaluate whether the current vitamin D content in ready‑to‑use therapeutic foods is sufficient for all children and whether additional supplementation improves clinical outcomes, especially in higher‑risk groups. Where deficiency is identified, children should be monitored and managed in accordance with local clinical guidelines and policies. Additionally, seasonality affects vitamin D levels, but it is unclear if vitamin D deficiency resolves with time or change in season. Longitudinal studies would clarify seasonal variation over time and its implications for post‑discharge care. Lastly, although no association was observed between vitamin D status and mortality, children identified as deficient at hospitalisation should still be closely monitored.
Supporting information
S1 Fig. Study schema.
Summary of the number of children living with SAM included in the 25(OH)D analysis in both Zambia and Zimbabwe at hospital discharge (Zambia: n = 101 and Zimbabwe: n = 341).
https://doi.org/10.1371/journal.pone.0355404.s001
(PDF)
S2 Fig. Plasma 25(OH)D concentrations at discharge by hospital of admission.
https://doi.org/10.1371/journal.pone.0355404.s002
(PDF)
S2 Table. Multivariable linear regression analysis, including all children.
https://doi.org/10.1371/journal.pone.0355404.s004
(PDF)
S3 Table. Multivariable linear regression analysis excluding children aged <6 months.
https://doi.org/10.1371/journal.pone.0355404.s005
(PDF)
S4 Table. Mean seasonal temperatures recorded in Lusaka and Harare, 1901–2020.
https://doi.org/10.1371/journal.pone.0355404.s006
(PDF)
S5 Table. Minimum seasonal temperatures in Lusaka and Harare, 1901–2020.
https://doi.org/10.1371/journal.pone.0355404.s007
(PDF)
S6 Table. Maximum seasonal temperatures in Lusaka and Harare, 1901–2020.
https://doi.org/10.1371/journal.pone.0355404.s008
(PDF)
S7 Table. Seasonal precipitation in Lusaka and Harare, 1901–2020.
https://doi.org/10.1371/journal.pone.0355404.s009
(PDF)
S8 Table. Dataset used for all analyses presented in this manuscript.
https://doi.org/10.1371/journal.pone.0355404.s010
(XLSX)
Acknowledgments
We acknowledge the entire study team involved in the HOPE-SAM study conducted in Harare and Lusaka. We also extend our gratitude to the caregivers and children who participated in the HOPE-SAM study.
References
- 1. Bhutta ZA, Berkley JA, Bandsma RHJ, Kerac M, Trehan I, Briend A. Severe childhood malnutrition. Nat Rev Dis Primers. 2017;3:17067. pmid:28933421
- 2.
WHO. Child growth standards and the identification of severe acute malnutrition in infants and children: a joint statement by the World Health Organisation and the United Nations Children’s Fund. Geneva: World Health Organisation; 2009. English. Available from: https://www.who.int/publications/i/item/9789241598163
- 3.
WHO. Child growth standards and the identification of severe acute malnutrition requiring inpatient care in children 6-59 months of age. World Health Organisation. Available from: https://www.who.int/tools/elena/interventions/sam-identification-inpatient
- 4. Asebe HA, Asmare ZA, Mare KU, Kase BF, Tebeje TM, Asgedom YS, et al. The level of wasting and associated factors among children aged 6-59 months in sub-Saharan African countries: multilevel ordinal logistic regression analysis. Front Nutr. 2024;11:1336864. pmid:38903623
- 5.
Zambia Statistics Agency, Ministry of Health (MoH) [Zambia], and ICF. Zambia Demographic and Health Survey 2024: Key Indicators Report. Lusaka, Zambia, and Rockville, Maryland, USA: Zambia Statistics Agency, MoH, and ICF; 2024. Available from: https://dhsprogram.com/pubs/pdf/PR159/PR159.pdf
- 6.
UNICEF. Preventing malnutrition in all forms; acute malnutrition has worsened since COVID-19, requiring treatment for severe wasting for 15,000 children. UNICEF; 2022. Available from: https://www.unicef.org/zimbabwe/reports/preventing-malnutrition-all-forms
- 7. Zyambo K, Hodges P, Chandwe K, Mweetwa M, Westcott J, Krebs NF, et al. Reduced fractional absorption of zinc in children with environmental enteropathy in Zambia. J Pediatr Gastroenterol Nutr. 2022;74(2):277–83. pmid:34469927
- 8.
Pocket book of hospital care for children: guidelines for the management of common childhood illnesses. 2nd ed. Geneva: World Health Organisation; 2013. Available from: https://www.ncbi.nlm.nih.gov/books/NBK154447/
- 9. Jones KDJ, Berkley JA. Severe acute malnutrition and infection. Paediatr Int Child Health. 2014;34(sup1):S1–29.
- 10. Ngari MM, Thitiri J, Mwalekwa L, Timbwa M, Iversen PO, Fegan GW, et al. The impact of rickets on growth and morbidity during recovery among children with complicated severe acute malnutrition in Kenya: a cohort study. Matern Child Nutr. 2018;14(2):e12569. pmid:29178404
- 11. Jones KDJ, Hachmeister CU, Khasira M, Cox L, Schoenmakers I, Munyi C, et al. Vitamin D deficiency causes rickets in an urban informal settlement in Kenya and is associated with malnutrition. Matern Child Nutr. 2018;14(1):e12452. pmid:28470840
- 12. Phiri TN, Mutasa K, Rukobo S, Govha M, Mushayanembwa P, Mwakamui S, et al. Severe acute malnutrition promotes bacterial binding over proinflammatory cytokine secretion by circulating innate immune cells. Sci Adv. 2023;9(44):eadh2284. pmid:37910623
- 13. Fenercioglu AK. The anti-inflammatory roles of vitamin D for improving human health. Curr Issues Mol Biol. 2024;46(12):13514–25. pmid:39727935
- 14. Aribi M, Mennechet FJD, Touil-Boukoffa C. Editorial: The role of vitamin D as an immunomodulator. Front Immunol. 2023;14:1186635. pmid:37056773
- 15. Maruotti N, Cantatore FP. Vitamin D and the immune system: Table 1. J Rheumatol. 2010;37(3):491–5.
- 16. Sanlier N, Guney-Coskun M. Vitamin D, the immune system, and its relationship with diseases. Egypt Pediatr Assoc Gaz. 2022;70(1):39.
- 17. Carboo JA, Malan L, Lombard MJ, Dolman-Macleod RC. Vitamin D status in relation to systemic and intestinal inflammation in undernourished children, 6–59 months old: Design and rationale of a non-controlled open label trial. Hum Nutr Metab. 2023;31:200181.
- 18.
UNICEF. RUTF Health Market Framework. United Nations Children’s Fund; 2020. Available from: https://www.unicef.org/supply/documents/healthy-market-framework-ready-use-therapeutic-food-rutf
- 19. Kelly P. The contribution of environmental enteropathy to the global problem of micronutrient deficiency. Proc Nutr Soc. 2021;80(3):303–10. pmid:33663621
- 20. Hasan MM, Gazi MA, Das S, Fahim SM, Hossaini F, Khan A-R, et al. Gut biomolecules (I-FABP, TFF3 and lipocalin-2) are associated with linear growth and biomarkers of environmental enteric dysfunction (EED) in Bangladeshi children. Sci Rep. 2022;12(1):13905. pmid:35974137
- 21. Regassa R, Tamiru D, Duguma M, Belachew T. Environmental enteropathy and its association with water sanitation and hygiene in slum areas of Jimma Town Ethiopia. PLoS One. 2023;18(6):e0286866. pmid:37352168
- 22. Amadi B, Zyambo K, Chandwe K, Besa E, Mulenga C, Mwakamui S, et al. Publisher Correction: adaptation of the small intestine to microbial enteropathogens in Zambian children with stunting. Nat Microbiol. 2022;7(1):183. pmid:34880416
- 23. Holick MF. Vitamin D status: measurement, interpretation, and clinical application. Ann Epidemiol. 2009;19:73–8.
- 24. Moreira CA, Ferreira CEDS, Madeira M, Silva BCC, Maeda SS, Batista MC, et al. Reference values of 25-hydroxyvitamin D revisited: a position statement from the Brazilian Society of Endocrinology and Metabolism (SBEM) and the Brazilian Society of Clinical Pathology/Laboratory Medicine (SBPC). Arch Endocrinol Metab. 2020;64(4):462–78. pmid:32813765
- 25. Ayadi ID, Nouaili EBH, Talbi E, Ghdemssi A, Rached C, Bahlous A, et al. Prevalence of vitamin D deficiency in mothers and their newborns in a Tunisian population. Int J Gynaecol Obstet. 2016;133(2):192–5. pmid:26952350
- 26. Sudfeld CR, Manji KP, Smith ER, Aboud S, Kisenge R, Fawzi WW, et al. Vitamin D deficiency is not associated with growth or the incidence of common morbidities among tanzanian infants. J Pediatr Gastroenterol Nutr. 2017;65(4):467–74. pmid:28644368
- 27. Abok II, Imoh LC, Bode-Thomas F, Oguche S, Zoakah A, Sagay A. Relationship between maternal and infant serum vitamin D levels in Jos, Plateau State, Nigeria: a cross-sectional study. Pan Afr Med J. 2023;46:48. pmid:38188884
- 28. Benameur T. Seasonal variations in 25-hydroxyvitamin D levels among pediatric patients attending the healthcare centre. Nutrients. 2024;16(3):379. pmid:38337664
- 29. Fontanive TO, Dick NRM, Valente MCS, Laranjeira VDS, Antunes MV, Corrêa MDP, et al. Seasonal variation of vitamin D among healthy adult men in a subtropical region. Rev Assoc Med Bras (1992). 2020;66(10):1431–6. pmid:33174939
- 30. Heidari B, Haji Mirghassemi MB. Seasonal variations in serum vitamin D according to age and sex. Caspian J Intern Med. 2012;3(4):535–40. pmid:24009930
- 31. Prentice A, Schoenmakers I, Jones KS, Jarjou LMA, Goldberg GR. Vitamin D deficiency and its health consequences in Africa. Clin Rev Bone Miner Metab. 2009;7:94–106.
- 32. Giustina A, Bilezikian JP, Adler RA, Banfi G, Bikle DD, Binkley NC, et al. Consensus statement on vitamin D status assessment and supplementation: whys, whens, and hows. Endocr Rev. 2024;45(5):625–54. pmid:38676447
- 33. Cui A, Zhang T, Xiao P, Fan Z, Wang H, Zhuang Y. Global and regional prevalence of vitamin D deficiency in population-based studies from 2000 to 2022: a pooled analysis of 7.9 million participants. Front Nutr. 2023;10:1070808. pmid:37006940
- 34. Gómez-Alonso C, Naves-Díaz ML, Fernández-Martín JL, Díaz-López JB, Fernández-Coto MT, Cannata-Andía JB. Vitamin D status and secondary hyperparathyroidism: the importance of 25-hydroxyvitamin D cut-off levels. Kidney Int Suppl. 2003;(85):S44-8. pmid:12753264
- 35. Bwakura-Dangarembizi M, Dumbura C, Amadi B, Ngosa D, Majo FD, Nathoo KJ, et al. Risk factors for postdischarge mortality following hospitalization for severe acute malnutrition in Zimbabwe and Zambia. Am J Clin Nutr. 2021;113(3):665–74. pmid:33471057
- 36. Bhat IB, Jan M, Bhat AS. Neurodevelopmental outcome of children with severe acute malnutrition. Int J Contemp Pediatr. 2023;10:238–40.
- 37. Sturgeon JP, Mufukari W, Tome J, Dumbura C, Majo FD, Ngosa D, et al. Risk factors for inpatient mortality among children with severe acute malnutrition in Zimbabwe and Zambia. Eur J Clin Nutr. 2023;77(9):895–904. pmid:37553508
- 38. Bwakura-Dangarembizi M, Amadi B, Bourke CD, Robertson RC, Mwapenya B, Chandwe K, et al. Health Outcomes, Pathogenesis and Epidemiology of Severe Acute Malnutrition (HOPE-SAM): rationale and methods of a longitudinal observational study. BMJ Open. 2019;9(1):e023077. pmid:30782694
- 39.
Pocket book of hospital care for children: guidelines for the management of common childhood illnesses. 2nd ed. Geneva: World Health Organisation; 2013. Available from: https://www.ncbi.nlm.nih.gov/books/NBK154447/
- 40. Bresson JL, Burlingame B, Dean T, Fairweather-Tait S, Heinonen M, Hirsch-Ernst KI. Dietary reference values for vitamin D. EFSA J. 2016;14(10).
- 41. EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA). Dietary reference values for vitamin D. EFSA J. 2016;14.
- 42. Bailey M, Heinrich D, Kruczkiewicz A. Climate profiles of countries in Southern Africa: Zimbabwe. Red Cross Climate Centre; 2019 [cited 2024 Jul 25]. Available from: https://www.climatecentre.org/wp-content/uploads/Climate-Profiles-of-Countries-in-Southern-Africa-Zimbabwe.pdf
- 43. Irene PV, Joseph S. Zambia: Climate and health; 2014 [cited 2024 Jul 25]. Available from: https://bpb-us-w2.wpmucdn.com/u.osu.edu/dist/9/1401/files/2014/03/Zambia-1gwih5z.pdf
- 44. Bourke CD, Jones KDJ, Prendergast AJ. Current understanding of innate immune cell dysfunction in childhood undernutrition. Front Immunol. 2019;10:1728. pmid:31417545
- 45. Nabeta HW, Kasolo J, Kiggundu RK, Kiragga AN, Kiguli S. Serum vitamin D status in children with protein-energy malnutrition admitted to a national referral hospital in Uganda. BMC Res Notes. 2015;8:418. pmid:26346815
- 46. Merker M, Amsler A, Pereira R, Bolliger R, Tribolet P, Braun N, et al. Vitamin D deficiency is highly prevalent in malnourished inpatients and associated with higher mortality: a prospective cohort study. Medicine (Baltimore). 2019;98(48):e18113. pmid:31770235
- 47. Ganmaa D, Bromage S, Khudyakov P, Erdenenbaatar S, Delgererekh B, Martineau AR. Influence of vitamin D supplementation on growth, body composition, and pubertal development among school-aged children in an area with a high prevalence of vitamin D deficiency: a randomized clinical trial. JAMA Pediatr. 2023;177(1):32–41. pmid:36441522
- 48. Middelkoop K, Micklesfield L, Stewart J, Walker N, Jolliffe DA, Mendham AE, et al. Influence of vitamin D supplementation on growth, body composition, pubertal development and spirometry in South African schoolchildren: a randomised controlled trial (ViDiKids). BMJ Paediatr Open. 2024;8:e002495.
- 49. Tisha SA, Ahmed SS, Mahmud S, Hasan MK, Islam S, Ratan MNH, et al. Effect of vitamin D supplementation in children with severe acute malnutrition. Glob Acad J Med Sci. 2023;5:12–20.
- 50. Saleem J, Zakar R, Zakar MZ, Belay M, Rowe M, Timms PM, et al. High-dose vitamin D3 in the treatment of severe acute malnutrition: a multicenter double-blind randomized controlled trial. Am J Clin Nutr. 2018;107(5):725–33. pmid:29722846
- 51. McIntyre S, Goldsmith S, Webb A, Ehlinger V, Hollung SJ, McConnell K, et al. Global prevalence of cerebral palsy: a systematic analysis. Dev Med Child Neurol. 2022;64(12):1494–506. pmid:35952356
- 52. Akpınar P. Vitamin D status of children with cerebral palsy (should vitamin D levels be checked in children with cerebral palsy?). North Clin Istanbul. 2018.
- 53. Paker N, Yavuz Mollavelioglu T, Bugdaycı D, Ones K, Bardak AN, Karacan I, et al. Vitamin D levels in children with cerebral palsy. J Pediatr Rehabil Med. 2023;16(1):163–9. pmid:36031913
- 54. Manohar S, Gangadaran RP. Vitamin D status in children with cerebral palsy. Int J Contemp Pediatr. 2017;4:615.
- 55. Alsoda M, Ali O, Kamel M. Evaluation of vitamin D status in children with cerebral palsy. J Med Sci Res. 2021;4:191.
- 56. Alenazi KA, Alanezi AA. Prevalence of vitamin D deficiency in children with cerebral palsy: a meta-analysis. Pediatr Neurol. 2024;159:56–61. pmid:39137591
- 57. Alenazi KA. Vitamin D deficiency in children with cerebral palsy: a narrative review of epidemiology, contributing factors, clinical consequences and interventions. Saudi J Biol Sci. 2022;29(4):2007–13. pmid:35531196
- 58. Bertolazi BS, Machado FR, Bálsamo EC, Segat HJ, dos Santos M, Boeira SP. Associations of socioeconomic, gestational, and vitamin D–rich food intake with vitamin D status in mothers of infants and young children. Discov Public Health. 2026;23(1):94.
- 59. Srivastava SB. Vitamin D: do we need more than sunshine? Am J Lifestyle Med. 2021;15:397–401.
- 60. Tian Y, Leng F, Zhou X, He Y, Li L, Ye R, et al. Prevalence and influencing factors of micronutrient powder adoption among children aged 6-24 months by parental and grandparental caregivers: an analysis from rural China. BMC Public Health. 2025;25(1):912. pmid:40055643
- 61. Saleh SE-S, Hendy A, Ibrahim RK, Badr EA. Maternal awareness and practices regarding vitamin D and their impact on health outcomes in children under five years in Egypt. Sci Rep. 2026;16(1):18869. pmid:42310028
- 62. Hays H, Flores LE, Kothari V, Bilek L, Geske J, Skinner A. Vitamin D status and seasonal variation: a retrospective single institution database study of patients pursuing metabolic/bariatric surgery. Clin Nutr Open Sci. 2022;41:1–9.
- 63. Grant WB, Bhattoa HP, Pludowski P. Determinants of vitamin D deficiency from sun exposure. In: Vitamin D. Elsevier; 2018. p. 79–90.
- 64. Hakeem MK, Hassan A, Rajendran T, Al-Menhali A, Yasin J, Gariballa S, et al. Harnessing natural sunlight indoors: sensor-regulated therapeutic approach to enhance vitamin D status in humans. Sci Rep. 2026;16(1):10723. pmid:41906059
- 65. Marondedze AK, Schütt B. Dynamics of land use and land cover changes in Harare, Zimbabwe: a case study on the linkage between drivers and the axis of urban expansion. Land. 2019;8(10):155.
- 66.
World Bank. Solar resource and PV potential of Zambia: Solar Resource Atlas. Washington (DC): World Bank; 2019 [cited 2024 Jul 25]. Available from: https://www.moe.gov.zm/wp-content/uploads/2022/06/Solargis_Solar_Resource_Atlas_Zambia_128-09-2019_WBG-ESMAP-1.pdf
- 67.
The World Bank. Climate Change Knowledge Portal: Zambia; 2021. Available from: https://climateknowledgeportal.worldbank.org/country/zambia/climate-data-historical
- 68.
The Zambia & Zimbabwe Food Region – Objective List. Available from: https://objectivelists.com/the-zambia-zimbabwe-food-region/
- 69. Watanabe K, Petri WA. Environmental enteropathy: elusive but significant subclinical abnormalities in developing countries. eBioMedicine. 2016;10:25–32.
- 70. Li H, Yuan S, Liao M, Tan S, Zheng J, Wan L, et al. Effects of HIV exposure on anemia and vitamin D nutritional status in children aged 6-24 months: a hospital-based cross-sectional study. Sci Rep. 2025;15(1):2839. pmid:39843726
- 71. Eckard AR, McComsey GA. Vitamin D deficiency and altered bone mineral metabolism in HIV-infected individuals. Curr HIV/AIDS Rep. 2014;11(3):263–70. pmid:24962286
- 72. Kruger MJ, Nell TA. Bone mineral density in people living with HIV: a narrative review of the literature. AIDS Res Ther. 2017;14(1):35. pmid:28747190
- 73. Eckard AR, Leong T, Avery A, Castillo MD, Bonilla H, Storer N, et al. Short communication: High prevalence of vitamin D deficiency in HIV-infected and HIV-uninfected pregnant women. AIDS Res Hum Retrovir. 2013;29(9):1224–8. pmid:23675655
- 74. Wierzbicka A, Oczkowicz M. Sex differences in vitamin D metabolism, serum levels and action. Br J Nutr. 2022;128(11):2115–30. pmid:35042577
- 75. Sanghera DK, Sapkota BR, Aston CE, Blackett PR. Vitamin D status, gender differences, and cardiometabolic health disparities. Ann Nutr Metab. 2017;70:79–87.
- 76. Johnson LK, Hofsø D, Aasheim ET, Tanbo T, Holven KB, Andersen LF, et al. Impact of gender on vitamin D deficiency in morbidly obese patients: a cross-sectional study. Eur J Clin Nutr. 2012;66(1):83–90. pmid:21792214
- 77.
Black RE, Laxminarayan R, Temmerman M, Walker N, editors. Reproductive, maternal, newborn, and child health: disease control priorities, vol. 2. 3rd ed. Washington (DC): The International Bank for Reconstruction and Development/ The World Bank; 2016. Available from: http://www.ncbi.nlm.nih.gov/books/NBK361907/
- 78. Walli NZ, Munubhi EK, Aboud S, Manji KP. Vitamin D levels in malnourished children under 5 years in a tertiary care center at Muhimbili National Hospital, Dar es Salaam, Tanzania-A Cross-sectional study. J Trop Pediatr. 2017;63(3):203–9. pmid:27794532
- 79. Heath AK, Kim IY, Hodge AM, English DR, Muller DC. Vitamin D status and mortality: a systematic review of observational studies. Int J Environ Res Public Health. 2019;16(3):383. pmid:30700025
- 80. Kumar MK, Das S, Biswal N, Parameswaran N, Nanda N. Vitamin D status at admission and its association with mortality in children admitted to the pediatric intensive care unit. Cureus. 2020.
- 81. Khamene SS, Khawajah IM, Moghtaderi M. Association between serum vitamin D levels and mortality in children receiving chronic dialysis: a retrospective cohort study. Health Sci Rep. 2026;9:e71804.