Dietary sources of sodium intake in nigerian adults: A population-based cross-sectional study

Abstract

Nigeria seeks to address the growing burden of hypertension and related diseases by reducing excessive dietary sodium through national dietary policymaking. This study aims to describe the levels and sources of dietary sodium intake among Nigerian adults to inform these policies. From June 2023 to July 2023, adults aged 18 to 69 years old were recruited from the Federal Capital Territory, Kano, and Ogun states to participate in a population-based, cross-sectional demographic health survey. Data were also collected to assess levels and dietary sources of sodium through four 24-h dietary recalls by trained study personnel. The primary analyses included the distribution of sodium intake and sources of sodium, in aggregate and by sex and state. Results were weighted to the Nigerian population. Multivariate regression models evaluated associations between baseline sociodemographic factors and sodium intake. Among 537 participants, 365 (68.0%) were female, median (IQR) age was 38 (27, 48) years, and 27.2% and 15.1% had a self-reported history of hypertension and cardiovascular disease, respectively. Most (90.7%) participants completed all 4 dietary recalls. Weighted median (IQR) daily sodium intake according to repeated 24-h dietary recalls was 3,876 (3,169, 4,783) mg per day with higher intake reported among males (3,832 [3,201, 4,658] mg/dl) compared with females (3,515 [2,859, 4,313], p < .0001). Nearly two-thirds (62.1%) of sodium came from discretionary sources, including 27.2% from salt and 32.5% from salty seasonings, 24.0% came from restaurant or street food, and 8.6% came from non-discretionary sources at home (i.e., sodium inherent in foods). Salt and salty seasonings added at the table accounted for 10.7% of sodium intake and was highest among females (21.6%) and males (16.2%) in Kano (p < .0001). On the other hand, sodium from street food was highest in males (35.7%) and females (34.2%) in Ogun. Older participants 60–69 years (adjusted beta [95% CI] = -332.8 mg (-639.0, -6.6) mg) had lower daily sodium intake compared to participants 30–44 years. Results were similar when excluding individuals with cardiovascular disease or hypertension. Adults in the Federal Capital Territory, Kano, and Ogun consume nearly twice the recommended level of dietary sodium. Most dietary sodium intake came from home cooked foods, nearly two-thirds of which were consumed from discretionary sources, which has important policy implications for dietary sodium policy implementation.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-23828-9.

Introduction

High dietary sodium intake is a significant public health concern worldwide, contributing to the global burden of non-communicable, chronic diseases (NCDs) such as hypertension, stroke, chronic kidney disease, and other cardiovascular diseases¹. Healthy diets, particularly diets low in sodium, play an important role in cardiovascular health, with excessive dietary sodium consumption being a major risk factor for hypertension and hypertension related complications^2,3. In Nigeria, as in many other countries, excessive dietary sodium consumption is prevalent based on national estimates and modeling studies^4–6. Reducing dietary sodium intake by 30% from 2019 to 2025 is a major target outlined in Nigeria’s 2019 National Multisectoral Action Plan for the Prevention and Control of Noncommunicable Diseases⁵ through implementation of the World Health Organization’s SHAKE (Surveillance, Harness, Adopt, Knowledge, Environment) technical package⁷.

While numerous studies have investigated sodium intake in population-based samples^1,8,9, including Nigeria⁶, fewer have evaluated the dietary sources of sodium¹⁰, and none, to our knowledge, among Nigerian adults. Understanding the sources of dietary sodium is crucial for developing effective interventions and policies to reduce excess dietary sodium intake and mitigate associated health risks¹¹. Given the diverse cultural and dietary practices within Nigeria, regional variation in sodium sources is also expected. Therefore, identifying these sources and their variability is essential for tailoring policies and corresponding multi-level strategies to promote healthy dietary habits and reduce excessive dietary sodium intake at the population level. Identifying sources and variability of dietary sodium intake is also recommended by the World Health Organization to guide evidence-informed policymaking¹.

The Nigeria Sodium Study includes three waves of retail surveys¹², stakeholder interviews¹³, and population surveys to evaluate the implementation and effectiveness of national dietary sodium policies in Nigeria. The aim of this study was to examine the sources of sodium in the diets among the first wave of a population-based sample of Nigerian adults from three distinct geographic regions: Federal Capital Territory (North Central), Kano State (Northwest), and Ogun State (Southwest). Results from the other waves will be reported in the future. States were identified and selected based on the goal to evaluate regional variability and previous collaborations among the study team, state-level investigators, and national and state public health officials. The study investigated the levels, sources, and differences in dietary sodium intake among Nigerian adults from these regions across baseline sociodemographic and clinical characteristics. By elucidating the sources of sodium in the diets of enrolled Nigerian adults across diverse geographic regions prior to national sodium policy implementation, this study will provide valuable baseline information for policymakers, public health professionals, and dietitians. These insights would aid in reducing sodium intake, promoting healthy dietary patterns in Nigeria, and ultimately decreasing the burden of diseases related to excessive dietary sodium intake through targeted interventions.

Methods

Population and recruitment

Recruitment was conducted among participants of the STEPwise approach to NCD risk factor Surveillance (STEPS) cross-sectional survey, which was led by the Nigerian Federal Ministry of Health and supported by the World Health Organization Nigeria office¹⁴. Participants were recruited using a multi-stage representative sampling frame and simple random sampling without replacement. In brief, 720 households from 60 clusters (12 per cluster) were selected to complete the STEPS survey, have anthropometry and seated blood pressure measured (the latter in triplicate), and provide biological samples in each of the 36 states in Nigeria plus the Federal Capital Territory. Additional participants were recruited to account for an anticipated 20% cumulative non-response rate.

From June 2023 to July 2023, a target sample of 450 adults aged 18 to 69 years old was recruited from 3 geographic locations (states): Federal Capital Territory (target n = 150), Kano (target n = 150), and Ogun (target n = 150) among STEPS participants. The study design, sample size, and methods for the dietary sources of sodium study was modeled after a previous study conducted in the United States, which included four repeated, multi-pass 24-h dietary recalls using trained and certified dietitians to estimate dietary sources of sodium intake¹¹. Additional participants were recruited to account for anticipated dropout as high as 20% across multiple dietary recall interviews. Participants who were pregnant or breastfeeding were excluded based on differences in dietary intake. Adult participants who were in the target age range and provided written informed consent were eligible. The institutional review boards at University of Abuja (2020/001/078), Northwestern University (STU00214999), and University of New South Wales (HC200807) approved the study procedures in accordance with the Declaration of Helsinki.

Overview of data collection

Data collection activities for the STEPS survey were performed by trained research staff from the Federal Ministry of Health, Public Health Departments of the 36 states and the Federal Capital Territory and World Health Organization Nigeria office, supported by Resolve to Save Lives. Data collection activities for the dietary sources of sodium study were performed by trained dietitians recruited by the Cardiovascular Research Unit of University of Abuja and University of Abuja Teaching Hospital following extensive training and certification overseen by the University of Minnesota Nutrition Coordinating Center. Data collection occurred in person in participants’ homes. Bar codes, unique to each participant, were used to ensure harmonization of participant data across multiple data sources.

STEPS survey

STEPS interviewers administered questionnaires to collect data on sociodemographic, health behaviors, medical history, and medication use. Height was measured in centimeters using a stadiometer (Seca, Germany), and weight was measured in kilograms using a digital scale (Seca, Germany). Body mass index was calculated by dividing each participants’ body weight in kilograms by the square of their height in meters (kg/m²). Hypertension and cardiovascular diseases were based on self-reports. Following the collection of spot urine on the first day visit, the laboratory scientists gave the participants instructions for an overnight fast and booked an appointment for the following day with the participants for fasting blood collection for blood glucose, total cholesterol, high density lipoprotein (HDL)-cholesterol, and triglyceride analysis. On the appointed day, a point-of-care device (Cardiocheck PA analyzer, manufactured by PTS Diagnostics, USA) was used to test a drop of capillary blood collected by finger prick, giving results for fasting blood glucose, total cholesterol, HDL cholesterol, and triglycerides.

Dietary sources of sodium

In addition to the questionnaire and biological samples provided in the STEPS survey, participants were consecutively invited to participate in this study. Participants were asked to complete 4 separate 24-h dietary recalls over the course of 1–2 weeks, including 1 recall on a weekend day, and to provide a 24-h urine sample. Participants were compensated ₦500 (~ 33 US Cents) for survey completion, ₦6000 (~ 4 US Dollars) for participation in 4 dietary recalls, and ₦3000 (~ 2 US Dollars) for 24-h urine collection.

Dietitians interviewed participants using the multiple-pass 24-h dietary recall method. Participants were provided verbal instructions and shown a standardized food atlas with pictorial representation of portioned meals, and household cooking measures, and an electronic weighing scale was used to estimate their amount of food intake during each recall. Responses provided by participants were documented by the dietitians in a 24-h dietary recall form. Recipes for foods prepared at home that had salt or salty seasonings as an ingredient were also collected from the household member who prepared the meal, including serving sizes, number of servings, and total yield (i.e., food prepared for the family). For every food item eaten, participants were asked to report the food source such as at home, restaurant, or street food, or other settings. Other settings were defined as meetings, events/occasions, parties, friends’ and relations’ homes, etc. For each meal, participants were asked if they added any salt or salty seasoning to their food at the table. If they had added salt or a salty seasoning at the table, then they were asked to estimate the amount added.

In conjunction with each diet recall, bagged salt samples ranging from 1 to 24 g were provided for participants to estimate the quantity of salt that was added in home cooked meals and at the table for each 24-h period. Participants were asked to select the bag that best matched the quantity of salt they added while cooking. Estimates were validated by requesting the household member who prepared the meal to estimate the quantity of salt added while cooking using a 1 kg sachet of salt that the dietitians provided as a reference. The actual amount of salt measured by the participants was then weighed on an electronic scale, especially in situations where the salt bags did not fit their exact description of the quantity used. In addition, the person who prepared the meals and household members were asked to estimate the quantity of bouillon and seasonings used in preparing foods, and the net weights of these were recorded and validated by weighing them on an electronic scale with a precision of 1 g (SF400, China).

24-Hour urine data

Participants were provided training to collect a 24-h urine sample by trained medical laboratory scientists to collect the sample approximately 3 days after initial enrollment. This approach is considered more reliable than spot urine sample collection¹⁵. Samples were collected from participants to overlap with one of the 24-h dietary recalls as much as feasible for comparison with estimated sodium intake. The urine samples were collected by the participants using a collecting jar, then poured into a five-liter container, and stored in a cooler bag with ice packs. Upon receipt by the laboratory scientist, the samples were transported in a cold box with ice packs to the regional certified laboratory at the Federal Capital Territory, Kano, and Ogun states, where 1.8 mL of each urine sample was aliquoted, stored at a 4° C refrigerator prior to ion-selective electrode analysis at International Organization for Standardization-certified laboratories in Abuja and Lagos, Nigeria within seven days.

Procedures for calculating sodium intake by source

The Nutrition Data System for Research (NDSR) software (version 2022), developed by the Nutrition Coordinating Center at the University of Minnesota of Minneapolis, MN, was used for calculating energy and sodium intake for the 24-h dietary recalls¹⁶. Prior to entering the dietary recalls into NDSR for nutrient calculation, recipes common to Nigeria were added to the software by local nutrition experts. Using data available in NDSR output files, sodium intake from the following were quantified for each 24-h dietary recalls: 1) salt or salty seasonings (e.g., bouillon) added to food during home preparation, 2) salt or salty seasonings added to food at the table for each location (home cooked, restaurant, street food, and others), and 3) sodium intrinsic in food or sodium content of commercially prepared foods for each location. Study team members visited restaurant kitchens to obtain recipes of commonly consumed meals; other recipes obtained were from street vendors (to determine their serving portions). The obtained recipes were analyzed to determine the frequently consumed meals, then standardized by local nutrition experts prior to entry of the dietary recalls into the NDSR for nutrient analysis, and the standardized (Nigerian) recipes were incorporated into the software. The standardized recipes were used to analyze the sodium content of unpackaged foods from restaurants, street vendors and others. The rules and procedures used to determine the proportion of sodium from each of these sources were based on previous research¹¹. To calculate sodium and potassium intake accurately for each participant, the dietary recall entry procedure included the following steps: (1) entering food source codes into the NDSR to identify the sources of sodium; (2) matching packaged foods and foods from the Nigerian Food Database 2019 to foods in the NDSR that were within 85 kcal and 50 mg of sodium per 100 g; (3) creating user recipes for foods that could not be matched to an existing food in the NDSR; (4) entering water data according to rules that reflected sodium levels depending on the water source; (5) coding extra salt added while eating with a user recipe named ‘Extra salt added while eating (NaSS001)’ and (6) entering household recipes when provided by participants in order to reflect the use of salt and other salty seasonings (e.g., bouillon) in home cooking. We considered variations during the 24-h recall interviews, which allowed the interviewer to enter personalized household recipes in the recipe section of the recall form. These recipes were further entered into the NDSR using the "assemble food recipe" option, providing an accurate reflection of what the participant consumed. Additionally, food measures and a food atlas were used to help estimate the total yield of the food and the portion sizes consumed by the participants. This entry was utilized to estimate their sodium intake, regardless of whether the recipe had been documented previously as a user recipe. Once dietary recalls were entered into the NDSR and reviewed with a multi-step quality assurance process, NDSR output data files were generated, which provided the sodium intake calculations with the source codes for analysis. These data and procedures also allowed us to estimate dietary potassium intake.

Statistical analysis

Primary analyses are reported in aggregate and by sex among participants with documented sociodemographic characteristics and at least one diet recall. Average nutrient intakes were calculated based on all dietary recalls collected for each participant for analyses. Continuous variables are reported as median with interquartile range (IQR) because most data were skewed. Differences in continuous variables by sex were assessed using Mann–Whitney U tests or Kruskal–Wallis tests, as appropriate. Categorical variables are reported as proportions. Differences in categorical variables by sex were assessed using Chi-Square tests. We explored variability in sodium (mg/day) or energy (calorie/day) intake between weekdays and weekend days, but did not observe significant differences; therefore, we did not weight the data accordingly. The 2022 World Bank male and female population estimates¹⁷ and the 2018 Nigeria Demographic and Health Survey¹⁸ were used to weight the sample analyses by sex, age, and rurality to estimate the population intake and excretion for adults aged 18–69 years.

We report the number of recalls completed and calculated the overall and median sodium intake by location in which foods were consumed. We further calculated the number and proportion of participants whose sodium intake exceeded the World Health Organization’s recommended daily limit (2,000 mg), the ratio of sodium intake compared with this limit, and the median energy intake (calories) and median sodium density (sodium [mg] per calorie [kcal]). We also calculated the proportion of participants whose potassium intake was below the World Health Organization (WHO) daily recommendation (3,500 mg) and the ratio of sodium intake compared with this level. The proportion of sodium intake from discretionary sources (salt and salty seasonings including bouillon, yaji spice, monosodium glutamate, and onga chicken seasoning, added at the table and during home cooking) and non-discretionary sources (sodium inherent in home-cooked, restaurant, street, and other foods) are also described. We will separately report sodium intake from packaged foods compared with sodium inherent in foods.

Urine samples from the Federal Capital Territory were standardized to 24 h but start and stop times were not collected for samples in Kano or Ogun so no adjustments were made for these. All samples were converted from mmol per day to milligrams per day based on urine volume. A 0.95 conversation factor was applied to account for incomplete excretion of sodium in urine based on the International Consortium for Quality Research on Dietary Sodium/Salt¹⁹. Some samples were excluded from our analysis based on implausibly low urine volume (< 300 ml) or creatinine excretion levels (< 6 mmol/day for males and < 4 mmol/day for females) for the primary analysis using previously published thresholds²⁰. The estimated caloric benchmark for sodium intake ranged from 1,000 kcal to 4,000 kcal. We conducted additional sensitivity analyses using alternative thresholds for 24-h urine collection completeness to evaluate the robustness of these results. We created Bland–Altman and difference-in-difference plots to evaluate agreement between 24-h urine sodium excretion and average diet recall sodium intake based on completeness as recommended by the International Consortium for Quality Research on Dietary Sodium/Salt²¹.

We evaluated average sodium intake across subgroup categories by sex, state, rurality (rural or urban), age (17–29 years, 30–44 years, 45–59 years, 60–69 years), level of educational attainment (no formal schooling, some schooling [no degree], high school completed, college or higher completed), marital status (never, current or cohabitating, formerly), religion (Christian, Muslim, other), smoking status (current versus not current), alcohol use in the last 30 days, body mass index (underweight or normal, overweight, obese), and blood pressure (< 140/90 mm Hg or ≥ 140/90 mm Hg) in bivariate tests and unadjusted regression analyses. Additionally, we evaluated average sodium intake across subgroup categories of history of blood pressure measurement, hypertension, cardiovascular disease, and diabetes, to consider the potential impact of dietary guidance that may accompany these clinical measurements and diagnoses. Bivariate analyses were repeated and stratified by state and rurality to explore regional variation.

Given the absence of significant clustering at both the state and cluster (i.e., geographic sampling unit) levels, indicated by an intraclass correlation coefficient of less than 5%, we conducted multivariable linear regression analyses with a generalized normal distribution to evaluate potential differences in sodium intake across groups. Generalized estimating equations were utilized to account for multiple diet recalls per participant. Interaction terms between state and sex were included in adjusted models if P < 0.10. This threshold was chosen to control for type II errors in detecting interactions that could be important for policy generation. The overall estimated mean sodium intake (mg/day) was reported for each state-sex combination in models including an interaction term and all other estimates represent the additive effects to these state-sex estimates associated with specific characteristics. Models were adjusted for sex, state, age, education, and, in an additional sensitivity model, energy intake (calories). History of ever having blood pressure or blood sugar measurements, and having ever been diagnosed with diabetes, were not adjusted for, as these factors were only significantly associated with sodium intake in unadjusted models. We also conducted an adjusted analysis to exclude participants without cardiovascular disease or hypertension to minimize the risk of reverse causality in subgroup analyses²². To ensure the robustness of the models, residual plots were examined for normality and adherence to heteroscedasticity. We conducted a complete case analysis and used a two-sided p < 0.05 to define statistical significance without adjustment for multiple testing, and a two-sided p < 0.10 to define statistical significance for interaction terms. Analyses were conducted using SAS version 9.4 (SAS Institute, Cary, NC).

Results

Figure 1 shows the flow of participant recruitment and analysis. Among 19,566 participants who completed the STEPS survey throughout Nigeria, 1,775 (9.1%) were from the Federal Capital Territory, Kano, or Ogun. Of these, 588 (33.1%) were approached for this study and 541 agreed to participate (92.0% of those approached). Four participants (0.1%) were excluded from analyses due to lack of sociodemographic data.

Figure 1.

Flowchart of participant selection and analysis. aConducted by the Nigerian Federal Ministry of Health and World Health Organization Nigeria office bIncludes Federal Capital Territory, Kano, and Ogun States cRandom sample of n=49 Clusters from the study catchment area; n=12 participants per cluster dDeclined (54.2%) or were unavailable (45.8%) to participate eDuplicate salt samples collected for salt added during home preparation and added at the table fUp to n=4 diet recalls per participant; at least n=1 recall on a weekend day gn=1 24-hour urine sample per participant Abbreviations: NDSR, Nutrition Data System for Research database; STEPS, STEPwise approach to NCD risk factor surveillance.

Table 1 shows the baseline sociodemographic and clinical characteristics in aggregate and stratified by sex. Among the included study participants (N = 537), 365 (68.0%) were females with a similar proportion of females recruited across the three states. Overall median (IQR) age was 38 (27, 48) years. Females had lower levels of education attainment (p = 0.004) and annual household income (p = 0.01), but a higher proportion of females were married or cohabitating (67.5% versus 60.2%) or formerly married (14.9% versus 4.1%, p < 0.0001) compared with males. A larger proportion of males were current smokers (10.5% versus 1.1%, p < 0.0001) and alcohol users (26.2% versus 10.1%, p < 0.0001) compared with females. Median (IQR) body mass index was higher among females compared with males (24.8 [21.2, 29.6] kg/m² versus 22.0 [19.9, 24.5] kg/m², p = 0.002), but median (IQR) systolic blood pressure was lower among females compared with males (122.3 [112.3, 137.3] mm Hg versus 128.7 [118.7, 139.7] mm Hg, p = 0.002). While 27.2% of participants who had a blood pressure measurement in the past self-reported having a history of high blood pressure, approximately one-third (32.4%) of all participants had high blood pressure (≥ 140/90 mm Hg) when measured at enrollment. Self-reported history of cardiovascular diseases (15.1%) was also similar between sexes. Baseline characteristics by state and rurality are reported in Supplemental Tables 1a and 1b. Briefly, participants from the Federal Capital Territory (40.8%) were more likely to be living in an urban area, have a higher education level and annual household income, higher body mass index, and higher diastolic blood pressure than the other states. Participants from rural areas (14.5%) had less education, lower annual household income, lower alcohol use, lower body mass index, and fewer had a history of blood pressure or blood sugar measurements.

Table 1.

Sociodemographic and Clinical Characteristics of Study Participants in Federal Capital Territory, Kano and Ogun States in Nigeria.

	Total N = 537 n (%)	Males n = 172 (32.0%) n (%)	Females n = 365 (68.0%) n (%)	P-Value
State				0.29
Federal Capital Territory	219 (40.8)	77 (44.8)	142 (38.9)
Kano	158 (29.4)	51 (29.7)	107 (29.3)
Ogun	160 (29.8)	44 (25.6)	116 (31.8)
Rurality				0.43
Rural	78 (14.5)	28 (16.3)	50 (13.7)
Urban	459 (85.5)	144 (83.7)	315 (86.3)
Age, median (IQR)	38 (27, 48)	38 (27, 49)	38 (27, 47)	0.57
Education level				0.004
No formal schooling	81 (15.1)	20 (11.6)	61 (16.7)
Some schooling (no degree)	160 (29.8)	40 (23.3)	120 (32.9)
High school completed	190 (35.4)	65 (37.8)	125 (34.3)
College or higher completed	47 (19.7)	47 (27.3)	59 (16.2)
Household Size (≥ 18 years), median (IQR)	2 (1, 3)	2 (1, 4)	2 (2, 3)	0.44
Annual HH income (Int$)a, median (IQR)	1,646 (633, 3,798)	2,304 (658, 4,254)	1,519 (532, 3,292)	0.01
Marital status				< 0.0001
Never married	125 (23.4)	61 (35.7)	64 (17.6)
Married or cohabitating	348 (65.2)	103 (60.2)	245 (67.5)
Formerly marriedb	61 (11.4)	7 (4.1)	54 (14.9)
Religion				0.31
Christian	239 (44.5)	69 (40.1)	170 (46.6)
Muslim	294 (54.7)	102 (59.3)	192 (52.6)
Other	4 (0.7)	1 (0.6)	3 (0.8)
Current smoker (at time of survey)	22 (4.1)	18 (10.5)	4 (1.1)	< 0.0001
Consumed alcohol in last 30 days	82 (15.3)	45 (26.2)	37 (10.1)	< 0.0001
Body Mass Index, kg/m2, median (IQR)	23.2 (20.4, 27.8)	22.0 (19.9, 24.5)	24.8 (21.2, 29.6)	< 0.0001
Systolic BP, mmHg, median (IQR)	124.7 (114.0, 139.0)	128.7 (118.7, 139.7)	122.3 (112.3, 137.3)	0.002
Diastolic BP, mmHg, median (IQR)	79.3 (73.0, 90.0)	80.8 (75.0, 89.8)	79.0 (72.0, 90.7)	0.33
High BP at enrollmentc	174 (32.4)	57 (33.1)	117 (32.1)	0.80
BP measured in the past	383 (71.3)	102 (59.3)	281 (77.0)	< 0.0001
History of hypertensiond	104 (27.2)	28 (27.5)	76 (27.1)	0.94
History of cardiovascular diseasee	81 (15.1)	23 (13.4)	58 (15.9)	0.45
Blood sugar measured in the past	174 (32.4)	47 (27.3)	127 (34.8)	0.08
History of diabetesf	23 (13.2)	6 (12.8)	17 (13.4)	0.91

Int$; (2022). One Int$ has the same purchasing power as 1 USD in the United States; the exchange rate was 755.29 ₦ = 1 USD in June 2023. Note: 27% missing.

^bIncludes separated, divorced, widowed.

^c ≥ 140/90 mm Hg.

^dParticipant was previously informed by a doctor of high blood pressure or hypertension (limited to participants who had a blood pressure measurement by a healthcare provider at least once).

^eParticipant reported previous heart attack, angina, or stroke.

^fParticipant was previously informed by a doctor of have raised blood sugar or diabetes (limited to participants who had a blood sugar measurement by a healthcare provider at least once).

BMI, body mass index; BP, blood pressure; HH, household; Int$, International dollar; IQR, Interquartile range.

Table 2 shows the daily sodium intake from the dietary recalls and 24-h urine sample in aggregate and stratified by sex. Most (90.7%) participants completed all 4 dietary recalls. Compared with females, males reported a higher calorie (2,396 versus 2,153 kcal/day, p < 0.0001) and sodium (3,832 versus 3,515 mg/day, p = 0.003) intake. When weighted to the Nigerian adult population distribution, the median (IQR) sodium intake was 3,876 (3,169, 4,783) mg/day. The proportion of average sodium intake across diet recalls that exceeded the WHO recommendation of sodium intake per day was similar between males and females (96.5% versus 94.5%, p = 0.32). Females reported a larger proportion of caloric (62.8%) and sodium (66.7%) intake inside the home compared to males (51.5%, p < 0.0001 and 55.9%, p < 0.0001, respectively) with the remainder of intake largely from street foods among both sexes. Among females, only 1.4% of sodium intake came from restaurants. Median (IQR) potassium intake, weighted to the Nigerian adult population, was 2,293 (1,854, 2,807) mg/day, with males having consumed significantly more than females (2,316 versus 2,082, p = 0.0002). Median (IQR) sodium excretion in 24-h urine, also weighted to the Nigerian adult population, was 3,170 (2,245, 4,189) mg/day and was similar between sexes. Assessments of 24-h urine collection completeness and correlation with dietary recalls are reported in Supplemental Table 3. Results stratified by state and rurality are reported in Supplemental Table 4a and 4b, respectively. In brief, sodium intake was highest in Kano, followed by Ogun and Federal Capital Territory (4,087 versus 3,523 versus 3,354 mg/day, p < 0.0001, respectively). A higher proportion of sodium intake came from the home in Federal Capital Territory (68.5%) and Kano (66.0%) compared to Ogun (52.6%, p < 0.0001). Furthermore, participants in rural communities reported higher sodium intake (4,046 versus 3,532 mg/day, p = 0.0002), with 95.2% of all participants exceeding the WHO recommended sodium intake.

Table 2.

Sources of Sodium Intake and Urinary Sodium Excretion among Study Participants in Federal Capital Territory, Kano and Ogun States in Nigeria.

	Nigeriaa	Total N = 537	Males n = 172 (32%)	Females n = 365 (68.0%)	P-Valueb
Recalls Per Participant, n (%)					0.80
1	-	8 (1.5%)	2 (1.2%)	6 (1.6%)
2	-	9 (1.7%)	3 (1.7%)	6 (1.6%)
3	-	33 (6.1%)	13 (7.6%)	20 (5.5%)
4	-	487 (90.7%)	154 (89.5%)	333 (91.2%)
Daily dietary recall
Sodium intake (mg/day)*	3,876 (3,169, 4,783)	3,608 (2,939, 4,434)	3,832 (3,201, 4,658)	3,515 (2,859, 4,313)	0.003
Proportion of total aggregate (sample-level) sodium intake by location of food source (95% CI)					< 0.0001
Home	62.9% (59.4, 66.4)	63.0% (59.9, 66.1)	55.9% (50.2, 61.6)	66.7% (63.1, 70.3)
Restaurant	2.4% (1.5, 3.2)	3.2% (2.1, 4.2)	6.7% (4.0, 9.4)	1.4% (0.7, 2.1)
Street	26.7% (23.1, 30.3)	24.7% (22.0, 27.4)	27.8% (22.8, 32.8)	23.1% (20.0, 26.3)
Othere	8.1% (6.5, 9.7)	9.1% (7.6, 10.6)	9.7% (7.0, 12.3)	8.8% (7.0, 10.6)
Participants in excessc, n (%)	97.1% CI95% = 95.7, 98.6	511 (95.2%)	166 (96.5%)	345 (94.5%)	0.32
Sodium excess ratio*d	1.9 (1.6, 2.4)	1.8 (1.5, 2.2)	1.9 (1.6, 2.3)	1.8 (1.4, 2.2)	0.003
Caloric intake (kcal/day)*	2,234 (1,883, 2,731)	2,221 (1,835, 2,694)	2,396 (2,039, 2,888)	2,153 (1,770, 2,588)	< 0.0001
Proportion of total aggregate (Sample-level) caloric intake by location of food source (95% CI)					< 0.0001
Home	57.8% (53.9, 61.7)	58.9% (55.7, 62.0)	51.5% (45.8, 57.1)	62.8% (59.1, 66.5)
Restaurant	2.3% (1.5, 3.1)	3.4% (2.4, 4.5)	6.9% (4.2, 9.5)	1.6% (0.9, 2.3)
Street	32.2% (28.0, 36.3)	28.9% (26.0, 31.8)	32.3% (27.1, 37.5)	27.1% (23.7, 30.5)
Othere	7.7% (6.1, 9.3)	8.8% (7.4, 10.2)	9.4% (6.8, 12.0)	8.5% (6.8, 10.2)
Sodium density (mg/kcal)*f	1.7 (1.4, 2.0)	1.6 (1.4, 1.9)	1.6 (1.4, 1.8)	1.6 (1.4, 1.9)	0.13
Home	1.9 (1.5, 2.4)	1.7 (1.4, 2.2)	1.7 (1.3, 2.2)	1.7 (1.4, 2.1)	0.53
Restaurant	1.5 (1.2, 1.9)	1.5 (1.0, 1.9)	1.5 (1.2, 1.8)	1.2 (0.6, 1.9)	0.06
Street	1.3 (1.0, 1.6)	1.3 (0.9, 1.6)	1.3 (0.9, 1.6)	1.3 (0.9, 1.7)	0.38
Othere	1.8 (1.4, 2.4)	1.7 (1.2, 2.3)	1.6 (1.2, 2.0)	1.7 (1.2, 2.3)	0.24
Potassium intake (mg/day)*	2,293 (1,854, 2,807)	2,186 (1,758, 2,732)	2,316 (1,990, 2,954)	2,082 (1,695, 2,644)	0.0002
Participants deficientg, n (%)	91.2% CI95% = 87.5, 94.9	492 (91.6%)	151 (87.8%)	341 (93.4%)	0.03
Potassium deficiency ratio*h	0.7 (0.5, 0.8)	0.6 (0.5, 0.8)	0.7 (0.6, 0.8)	0.6 (0.5, 0.8)	0.0002
Daily (24-hour) urine
Sodium excreted (mg/day)*i	3,170 (2,245, 4,189)	3,006 (2,089, 4,269)h	3,069 (2,189, 4,261)i	3,004 (2,027, 4,294)j	0.91
Participants in excessc, n (%)	78.9% CI95% = 72.2, 85.6	275 (77.0%)	82 (78.9%)	193 (76.3%)	0.60

* Median (IQR).

^aStudy results weighted to reflect the Nigerian adult population (age 18–69) distribution by sex, age, and rurality.

^bMann–Whitney or Chi-square comparison of males versus females.

^cNumber of participants with mean daily sodium intake exceeding the WHO recommendation of < 2,000 mg/day.

^dSodium intake (mg/day) divided by the maximum WHO recommended intake (2,000 mg/day).

^eCaptured as meetings, events/occasions, parties, friends’ and relations’ homes, etc.

^fSodium (mg) per calorie (kcal).

^gNumber of participants with mean daily potassium intake falling below of the WHO recommendation of ≥ 3,500 mg/day.

^hPotassium intake (mg/day) divided by the minimum WHO recommended intake (3,500 mg/day).

ⁱAmong samples ≥ 300 ml and creatinine ≥ 6 mmol/day for males and ≥ 4 mmol/day for females (^hn = 357; ⁱn = 104; ^jn = 253).

IQR, interquartile range; WHO, World Health Organization.

Figure 2 shows the proportion of sodium intake by location of food source, stratified by sex and state. Most (62.1%, IQR: 45.1, 76.3) sodium was derived from discretionary sources, including salt added during home cooking and at the table. Salt added during cooking was the largest contributor (25.7%), followed by bouillon in all forms (22.7%). Sodium from salty seasonings added during home food preparation was highest among females in the Federal Capital Territory and contributed to 42.6% of total sodium intake. Salt and salty seasonings added at the table, primarily yaji spice which includes salt and bouillon (data not shown), was highest among females (18.5%) and males (14.2%) in Kano State. On the other hand, sodium from street food was highest in males (35.9%) and females (34.2%) in Ogun State.

Figure 2.

Proportion of Sodium Intake, mg/day, by Location of Food Source. aAdded at the Home, Restaurant, Street, or Other Table bIncludes Bouillon, Yaji Spice, Monosodium Glutamate (MSG), Onga Chicken Seasoning cCaptured as meetings, events/occasions, parties, friends’ and relations’ homes, etc. dStudy results weighted to reflect the Nigerian adult population (age 18–69) distribution by sex, age, and rurality eSodium (mg) per calorie (kcal).

Figure 3 shows median daily sodium intake by subgroups and stratified by sex. Among males, there were differences in unadjusted sodium intake by state, religion, and alcohol use. Among females, there were differences in unadjusted sodium intake by state, rurality, level of education, religion, history of blood pressure measurement, and history of diabetes.

Figure 3.

Median [IQR] Dietary Sodium Intake (mg/day) Reported by Diet Recall, by Sex. NOTE: P-values for bivariate tests (Mann-Whitney or Kruskal-Wallis) comparing sodium intake across categories. Abbreviations: BMI, body mass index, kg/m2; BP, blood pressure; CVD, cardiovascular disease.

The results of the unadjusted and adjusted regression models investigating sodium intake from diet recalls across sociodemographic characteristics are shown in Table 3. After adjustment for age and education, the difference in daily sodium intake between females and males by state was statistically significant (state-sex interaction terms p_Kano = 0.05 and p_Ogun = 0.03 versus Federal Capital Territory). The highest daily sodium intake was among males in Kano (4,267 mg [95% CI: 3,942 mg, 4,592 mg]), and the lowest daily sodium intake was among females in the Federal Capital Territory (3,274 mg [95% CI: 3,048 mg, 3,500 mg]). Participants aged 60–69 years old had −336.1 mg (95% CI: −650.9 mg, −21.4 mg) lower daily sodium intake compared with participants 30–44 years old (p = 0.04). Calorie intake was also associated with higher sodium intake (β = 1.10 [95% CI: 0.99, 1.21]). Results of sensitivity analyses adjusted for calories and sodium excretion in 24-h urine and subgroup analyses among those without self-reported history of cardiovascular disease and without cardiovascular disease or hypertension are reported in Supplemental Table 5 In brief, sensitivity and subgroup analyses showed a similar direction and magnitude of associations after multivariate adjustment.

Table 3.

Sociodemographic and Clinical Predictors of Dietary Sodium Intake among Study Participants in Nigeria.

	Unadjusted models		Adjusted modela
	β (95% CI)	P-Value	β (95% CI)	P-Value
Intercept	-	-	3,882.6 (3,611.1, 4,154.1)	-
Female (versus Male)	−293.3 (−484.4, −102.3)	0.003	−608.8 (−886.9, −330.6)	< 0.0001
State (versus federal capital territory)
Kano	709.6 (500.7, 918.4)	< 0.0001	384.8 (21.4, 748.2)	0.04
Ogun	218.1 (10.0, 426.2)	0.04	−104.4 (−489.7, 281.0)	0.60
State – sex interaction (versus Federal capital territory – female)
Kano—female	-	-	429.5 (−4.0, 863.0)	0.05
Ogun—female	-	-	491.9 (48.2, 935.5)	0.03
Age, years (versus 30–44)b
18–29	108.5 (−107.5, 324.4)	0.32	45.3 (−161.6, 252.3)	0.67
45–59	−206.2 (−448.8, 36.5)	0.10	−167.7 (−401.3, 65.8)	0.16
60–69	−232.7 (−549.3, 83.8)	0.15	−336.1 (−650.9, −21.4)	0.04
Education level (versus high school completed)b
No formal schooling	30.8 (−243.1, 304.6)	0.83	−58.0 (−348.5, 232.5)	0.70
Some schooling (no degree)	90.7 (−130.7, 312.2)	0.42	107.9 (−111.6, 327.4)	0.34
College or higher completed	−311.3 (−561.5, −61.1)	0.01	−209.0 (−451.1, 33.0)	0.09
Caloriesc	1.10 (0.99, 1.21)	< 0.0001	-	-

Mean Sodium Intake (95% CI), mg/day, by State and Sex, adjusted for age and education^d.

^aAll variables included in the table are adjusted for in the regression model.

^bAdditive effect to State – Sex Estimates.

^cPer 1 Calorie (kcal), Centered at 2,000 kcal/day.

^dMean estimated sodium from adjusted model {adjusted for age (30–44 years old) and education (High school completed)}.

β, beta coefficient; CI, confidence interval; FCT, Federal Capital Territory.

Discussion

This study used rigorous multi-pass dietary recall methods, along with concurrent 24-h urine collection, to estimate sodium intake and sources of sodium among Nigerian adults in the Federal Capital Territory, Kano State, and Ogun State. Results demonstrate that adjusted median dietary sodium intake was nearly two-fold higher than the WHO daily recommended level of 2,000 mg per day^1,7. Dietary sodium intake was higher among males compared with females due to higher calorie intake. Sodium intake was also higher in Kano compared with the other two states, likely due to cultural practices^23,24. For example, in Kano State, it is common to enjoy ready-to-eat meals with yaji, a spicy condiment. In rural areas, the yaji mixture typically consists of dried red chili pepper, ajinomoto seasoning, and salt. In contrast, urban areas often use a different blend, which includes red chili pepper, salt, ground ginger, and bouillon cubes. Additionally, Kano State is renowned for its tradition of community eating. In contrast with results from high-income countries like the United States¹¹ and consistent with Nigerian cultural practices, discretionary sources were the leading source of dietary sodium and calorie intake, led by salt added during preparation and bouillon. These results also demonstrated a relatively high proportion of dietary sodium from street foods, which are widely available and affordable, yet are also an underrecognized and largely unregulated source²⁵. Dietary potassium intake was also lower than recommended by the World Health Organization and may be attributed to poor dietary diversity among participants²⁶.

Geographic variability in the dietary sources of sodium has been reported. For example, in a 2020 systematic review of 80 dietary sources of sodium studies conducted in 34 countries, there was an inverse correlation between per capita gross domestic product and discretionary salt use¹⁰. In this review, more than half of sodium intake comes from discretionary salt in low- and middle-income countries, which aligned with the current study’s results. However, none of these studies were from Nigeria, the most populous country in Africa, with the only other sources of sodium study from Africa conducted in Mozambique among 100 hospital workers. In the Mozambique study, discretionary salt intake represented 60% of the daily sodium intake, which was estimated to be 4,220 (SD: 1,830) mg per day based on 24-h urinary excretion²⁷, which is also similar to this study.

Previous epidemiological research in Nigeria reported lower estimates of dietary sodium intake. For example, Odili et al. reported a median (IQR) urinary sodium excretion of 1,783 (1,892) mg/d among 2,509 adults in 12 Nigerian communities⁶. However, variability was wide in this study, and detailed dietary data were not collected to corroborate these urinary findings, which likely explains the differences observed in the current study. Further, the 2023 World Health Organization report on dietary sodium intake estimated that daily sodium intake in Nigeria was 2,524 mg per day (95% uncertainty interval: 2,376, 2,673) based on modeling data¹. However, dietary sodium intake has been consistently underestimated in low- and middle-income countries in large modeling studies when compared with population-based sampling studies like this one²⁸. On the other hand, the 2019 baseline reported in the National Multisectoral Action Plan for the Prevention and Control of Noncommunicable Diseases was higher at 4,000 mg of sodium, or 10 g of salt (1 and 3/4 teaspoons), per day⁵, compared with approximately 3,876 mg of sodium, or 9.5 g of salt (1 and 2/3 teaspoon) per day in this study. Thus, the results likely provide a closer estimate of the true dietary sodium intake level in Nigeria than previous studies^6,28.

These results have important policy implications as Nigeria seeks to implement the World Health Organization SHAKE technical package through the National Multisectoral Action Plan for the Prevention and Control of Noncommunicable Diseases⁵. Policy priorities outlined in the Action Plan include mandatory sodium limits in packaged foods, front-of-package labeling, and mass media and education campaigns to raise awareness about healthy diets, which align with the SHAKE package. These policies are projected to be cost effective²⁹ and are considered appropriate by many stakeholders, especially given the high sodium levels and incomplete sodium labeling on packaged foods in Nigeria¹². On the other hand, barriers to SHAKE package policy implementation include low awareness of the harms of excess sodium intake, high cost of healthy foods, and taste preference for higher sodium foods¹³. In contexts where discretionary salt use is high, like Nigeria, additional policy strategies should be considered to better match the sources of sodium through multisectoral stakeholder engagement³⁰. For example, potassium-enriched salts are effective, cost saving strategies to reduce blood pressure, total mortality, cardiovascular disease mortality, and cardiovascular disease events^31,32. Potassium-enriched salts could be promoted or subsidized, including in high-sodium products such as bouillon, as policy options to complement other SHAKE package strategies. Health education and mass media strategies may help to increase knowledge and awareness about discretionary sources of excess dietary sodium consumption and the role of potassium-enriched salts, among other evidence-based interventions to reduce excess dietary sodium intake.

This study included numerous strengths, including a large, population-based sampling frame of both sexes in a setting where dietary sources of sodium have not been rigorously assessed. Additional strengths include training and certification of dietitians for high quality data collection, adaptation of the NDSR for the Nigerian context, multiple pass dietary data collection with concurrent 24-h urine collection on both weekdays and weekends, evaluation of location of food source of sodium, and a high proportion of repeated dietary recalls collected. These strengths enhance the rigor, reproducibility, and reliability of the study findings.

This study also has some limitations. First, participation among females was higher than among males, which may be due to the household recruitment approach. Therefore, given the observed sex differences in dietary sodium intake, the study sample may underestimate the true dietary sodium intake on the population level. Second, data were collected in only three out of the 36 states in Nigeria, which may limit the generalizability of the findings. However, these states are geographically distributed through Nigeria, and we observed significant differences between states that may be relevant to those respective regions. To mitigate both limitations, we weighted the results based on the Nigerian adult population. Third, data were collected in a two-month period and may have missed longer-term seasonal trends. On the other hand, this study is the first of three waves of population surveys to estimate dietary sodium intake in Nigeria, and future results may refine these results. Fourth, results from 24-h urine collection were generally lower than those from dietary recalls. This may be due, at least in part, to under-collection of urine samples, including in Kano and Ogun States where start and stop times were not collected. However, we conducted sensitivity analyses using numerous published thresholds for urine volume and creatinine and adjusted for insensible losses as recommended¹⁹. Notably, in the Federal Capital Territory, where we standardized sodium excretion levels to 24 h, the 24-h urine sodium excretion closely aligned with dietary intake in males and exceeded dietary intake in females. Despite the lack of standardization in Kano and Ogun, we still show high levels of agreement based on reporting recommendations²¹.

Conclusions

Adults in the Federal Capital Territory, Kano State, and Ogun State consume nearly two times the recommended level of dietary sodium recommended by the World Health Organization, largely from home cooked foods, nearly two-thirds of which comes from discretionary sources. This excessive intake of dietary sodium, coupled with low dietary potassium intake, contributes to the large and growing burden of hypertension and hypertension-related diseases in Nigeria, underscoring the need for targeted interventions. These findings have important policy implications for the implementation of the SHAKE package, including the potential benefits of promoting the use of potassium-enriched salts and bouillon.

Abbreviations

IQR

Interquartile range

NCD

Noncommunicable chronic disease

NDSR

Nutrition data system for research

SHAKE

Surveillance, harness, adopt, knowledge, environment

SD

Standard deviation

STEPS

STEPwise approach to NCD risk factor surveillance

Author contributions

DBO, BN, MDH conceived the study. CEO, ANM led dietary data collection with training and certification conducted and overseen by JMLP and LJH. CO, MAO, AC, FA, DO, MMT, AS, GA, AFF, UU, DJ, RFA led STEPS survey data collection and EJ conducted the statistical analyses. All other authors (AO, LVH, VA, AIO, HE, RO, GI, LRH, AJ, KT, MM, MI, VT) contributed to study design, execution, interpretation of results, edits, and final approval.

Funding

The study was supported by the National Heart, Lung, and Blood Institute (NHLBI, UH3HL152381). Except for feedback from the NHLBI Project Scientist and Clinical Trial Specialist, the funder was not involved in the development of the study design; collection, management, analysis, and interpretation of data; writing of the report; and the decision to submit the report for publication.MDH has received travel support from the World Heart Federation and consulting fees from PwC Switzerland. MDH has an appointment at The George Institute for Global Health, which has a patent, license, and has received investment funding with intent to commercialize fixed-dose combination therapy through its social enterprise business, George Medicines. MDH has pending patents for heart failure polypills. The other authors have no relationships with industry to disclose.

Data availability

De-identified data, data dictionary, and statistical code will be made available through BioData Catalyst and by requests to Dr. Dike Ojji from the University of Abuja (dike.ojji@uniabuja.edu.ng).

Declarations

Competing interests

MDH has received travel support from the World Heart Federation and consulting fees from PwC Switzerland. MDH has an appointment at The George Institute for Global Health, which has a patent, license, and has received investment funding with intent to commercialize fixed-dose combination therapy through its social enterprise business, George Medicines. MDH has pending patents for heart failure polypills. The other authors have no relationships with industry to disclose.

Consent to participate

All participants provided written informed consent.

Ethics approval

The institutional review boards at University of Abuja, Northwestern University, and University of New South Wales approved the study procedures in accordance with the Declaration of Helsinki.

Disclaimer

The views expressed in this manuscript are those of the authors and do not necessarily represent the views of the National Heart, Lung, and Blood Institute; the National Institutes of Health; or the U.S. Department of Health and Human Services.