Introduction

Cardiovascular diseases (CVD) account for an estimated 28.1% of the total deaths in India. In 2016, 1.63 million deaths were attributable to hypertension compared to 0.78 million deaths in 19901. A diet high in sodium could contribute to 70 million Disability Adjusted Life Years (DALYs’) and 3 million deaths globally because of its association with CVDs2. A high sodium level in the diet also increases the risk of stroke and cardiac failure severity3,4. Besides the harmful cardiovascular effects of salt consumption, it could also be a potential risk factor for gastric cancer5,6. Restricting daily salt intake to the World Health Organization (WHO) recommended level of up to 5 g for adults is a beneficial and cost-saving way to achieve a 25% reduction in the prevalence of raised blood pressure and a 30% reduction in mean population salt intake by 2025. India’s National NCD Monitoring Framework has defined a similar percentage reduction, adapted from WHO’s voluntary global NCD targets7,8.

The WHO urges its Member States to strengthen dietary salt reduction strategies9. The benefits of attaining the WHO’s objective of reducing daily salt consumption to 5 g per adult by 2030 are substantial. This achievement is estimated to prevent an additional 87,870 premature ischemic heart disease (IHD) cases by 2050 and avoid 126,010 premature strokes10. Realising this target would generate an estimated £1260 million in extra healthcare savings alongside these health improvements. While the rationale for salt reduction provided by scientific evidence is robust, the quantum of available data is insufficient to translate such scientific insights into policies for reducing population salt intake. Preventing dietary salt intake must be driven by perceptions, behaviours, and practices influenced by diverse socio-economic and socio-cultural factors. Studies on the assessment of salt intake in India have been done in limited geographical settings; hence the result findings may not apply at a national level11,12,13. The lack of comprehensive data draws attention to the need for a national-level monitoring system to generate evidence on dietary salt intake, knowledge, attitude and practices. Feasible and standard methods of salt estimation incorporated into surveys helps monitor and plan better. These include 24-h urine measurement and spot urine measurement14. Although the 24-h urine collection method is acknowledged as the ‘gold standard’ approach for assessing salt intake, collecting urine over a whole day is laborious, costly, inconvenient, and intricate for the individuals involved. Spot urine has been widely used and validated to estimate dietary sodium15,16,17,18.

The Indian Council of Medical Research—National Centre for Disease Informatics and Research, Bengaluru undertook a comprehensive National NCD Monitoring Survey (NNMS) to establish a monitoring system that addresses the progress toward achieving the national NCD targets19. This study aimed to estimate the mean dietary salt intake and assess the awareness, behaviour and determinants of salt intake in an adult population of 18–69 years.

Materials and methods

Survey setting, population and sampling

The National NCD Monitoring Survey (NNMS) 2017–18 was conducted on a sample of 12,000 adults in the 18 to 69 age group to generate national-level key NCD-related indicators identified in the national NCD monitoring framework, the details of which are published elsewhere20. Urinary sodium excretion in spot urine samples was estimated in a subsample of 3000 adults, selected through simple random sampling in 150 nationally representative Primary Sampling Units, of which 75 were in urban, and the other 75 were in rural areas. Data was collected regarding awareness of the adverse health effects of high salt intake and the importance of lowering dietary salt intake. ‘Behaviour’ alluded to actions regarding salt intake in the daily diet. The sociodemographic, behavioural and metabolic determinants of dietary salt intake estimated from spot urinary samples were also analysed.

Data and sample collection

The study instrument for the survey was developed from reliable resources related to NCD risk factor surveillance21,22,23. Study participants were recruited after written and verbal consent. Qualified and trained medical social workers collected details at the household, individual and health facilities. Data on behaviours, perceptions and attitudes related to salt use were collected through face-to-face interviews. Camps were organised to conduct anthropometric, blood pressure and blood glucose measurements and collection of urine samples. A day before the camp, consenting participants were given 30 mL wide-mouthed labelled sterile urine collection bottles with a barcode specific to a selected adult to avoid sample mismatch. The spot urine sample was gently shaken to ensure homogeneity and transferred to 2 mL collection vials using Pasteur’s pipettes. The vials were then placed in a 9 × 9 storage box in a vaccine carrier containing ice packs for safe transportation at 4 °C to pre-appointed storage facilities, stored at − 20 °C till further shipment to a reference laboratory at the Department of Cardiac Biochemistry, All India Institute of Medical Sciences, New Delhi. The samples were tested in batches of 80 to 100 each.

Estimation of urinary sodium

The indirect Ion-Selective Electrode (ISE) method was used to estimate urinary sodium (Na) and potassium (K) levels using automated analysers (AU680 Chemistry analyser, Beckman Coulter, CA, USA)24. The ISE method has several advantages: it is faster, more accurate, and has lesser sample requirements. The measurable range is 20–200 mmol/L with a precision of < 0.1%. Internal Quality Control (IQC) samples in normal and pathological ranges were run for each batch of urine samples. Jaffe’s method was employed for urinary creatinine level estimation using Roche analyser (P800 Modular Analytics, Roche Diagnostics, Mannheim, Germany) through a commercial kit (Ref. 11,875,418–216, Roche diagnostics, Germany). The dietary salt intake from the spot urine sample was estimated using the INTERSALT (International Cooperative Society on Salt and Blood Pressure) equation with potassium for each sex25. The value obtained in mmol/L was multiplied by 2.54 to derive the daily mean salt intake in grams per day.

Data analysis

Data were cleaned using the IBM Statistical Package for Social Sciences (SPSS) windows—version 22.0. The statistical analysis using weighted data was carried out in STATA 14.1 using a complex survey analysis method26. Descriptive analysis was used to present results on the behaviour, perception and practices related to salt intake. The data distribution was checked using the graphical representation and then tested for normality with the Shapiro–Wilk test. Mean weighted salt intake comparison between the subgroups (sociodemographic, rural/urban, behavioural, physiological and metabolic risk factors) was made using the Student’s independent samples t-test and Analysis of Variance (ANOVA) test. Logistic regression analysis was carried out to determine the relationship between independent variables, and those variables with a p-value < 0.25 were taken in the multivariate analysis27. The final multivariate-adjusted model was fitted by adjusting the variables which included age, gender, area of residence, educational status etc.

Ethics approval

This study was approved by the Institutional Ethics Committee of the Coordinating Centre, ICMR-NCDIR. Approval no: NCDIR/IEC/2017/4 dated 03 February 2017. Before starting the survey, every implementing agency obtained its ethics approval from its Institutional Ethics Committee. Informed consent was obtained from all the study participants. As a post-research benefit, trained social investigators counselled the study participants to reduce salt consumption. They were also provided with a brochure describing tips for a healthy lifestyle in the local language. Since the research involved human participants, it was conducted per the Declaration of Helsinki.

Results

10,659 adults aged 18–69 years participated in the survey (response rate of 96.3%). Among the subsample of adults selected for the urinary sodium excretion study, 2643 study participants provided spot urine samples, of which 2266 (85.7%) samples were the final number of samples processed and included for analysis. The remaining 377 either had incomplete interviews or the urine samples were contaminated. A flow diagram that indicates the  inclusion of study participants is shown in the supplementary file.

Awareness and behaviour of the population on dietary salt intake

The proportion of adults who knew that daily high salt intake could affect health was less than one-third among respondents in all age categories, gender, rural–urban areas and those without education. However, over half of the participants were familiar with the importance of lowering salt intake in the diet. The percentage of adults who used extra salt in food or consumed too many salt-containing products was less than a quarter in all the sub-categories of sociodemographic, behavioural and metabolic factors (Table 1).

Table 1 Perceptions and behaviours on dietary salt intake.
Full size table

Weighted mean salt intake estimates according to the sociodemographic profile of the study population

A salt intake of > 5 g/day was observed across all the strata, as shown in Table 2. The estimated mean daily salt intake was 8.0 g (8.9 g/day for men and 7.1 g/day for women). The daily mean salt intake was significantly higher (p < 0.0001) in men (8.9 g), employed (8.6 g), current tobacco users (8.3 g), obese (9.2 g) and those with raised blood pressure (8.5 g). The population characteristics and their weighted mean dietary salt intake estimates in India from spot urinary samples according to age groups (18-44 and 45-69 years) and gender are shown in the Supplementary text.

Table 2 Population characteristics of adults aged 18–69 years and their weighted mean dietary salt intake estimates in India from spot urinary samples (n = 2266).
Full size table

Association between sociodemographic, behavioural, physiological and metabolic factors with salt intake

A salt intake of > 5 g per day was significantly higher in men than women [Unadjusted OR = 6.69 (2.52–17.72)]; [Adjusted OR = 17.66 (5.24–59.46)], among rural participants than in urban [Adjusted OR = 6.14 (1.83–20.60)], overweight and obese persons [Unadjusted OR = 10.95 (2.24–53.50); Adjusted OR = 17.62 (3.17–98.07)]. The findings are presented in Table 3.

Table 3 Association between demographic, behavioural and metabolic risk factors with salt intake (g/day).
Full size table

Consumption of high salt containing foods among adults

Approximately 43.8% of adults indicated that they consumed high-salt homemade food items every week (1–6 days per week), while 36.1% reported consuming salty snacks like namkeen, papad, and packaged chips at least once a month (1–3 days or less than once a month), as seen in Fig. 1.

Figure 1
figure 1

Frequency of consumption of high salt containing foods among adults.

Full size image

Discussion

The present study provides a national-level description of the awareness and behaviour of dietary salt intake and its determinants in terms of the sociodemographic (age, sex, area, education and employment), behavioural (current tobacco and alcohol use, physical activity and yoga practice), physiological (body mass index, blood pressure) and metabolic factors (blood glucose and cholesterol level) in the Indian population. While most respondents knew the importance of reducing dietary salt consumption, the proportion of those who perceived that a daily high salt intake could affect one’s health was low. The pattern of perception and behaviour did not substantively vary according to sociodemographic, behavioural and metabolic factors. These findings are similar to the survey conducted in Delhi, Haryana, and Andhra Pradesh, in which salt intake-related awareness and behaviour did not vary across different levels of education28. Studies done in specific and geographically limited settings have shown varying results. Close to two-thirds (64%) of the participants residing in an urban slum in Chandigarh, North India, did not consider lowering the salt intake important29. In another community-based study in North India, 46·6% from rural Haryana and 45·1% from urban Delhi were aware of the need to reduce dietary salt content30. In our study, it was positive to note that less than a third of the respondents affirmed that they added extra salt to their food or consumed far too many salty items. However, less than half of the study participants practised measures to control dietary salt intake. The most commonly adopted step was avoiding meals outside of the home (Supplementary Table S1).

In the present study, the weighted mean salt intake was more than the WHO-recommended salt intake of up to 5 g per day across all groups categorised according to the sociodemographic, behavioural and metabolic factors groups. Similar findings have been reported in other studies done in India. A systematic review of the mean salt intake in India found the observed value of 11 g per day31. Analogous study findings for mean dietary salt intake were reported in Andhra Pradesh (8.72 g/day) and Delhi, Haryana (5.62 g/day), which was calculated from the average estimates from two 24-h dietary recall surveys. The Chennai Urban Rural Epidemiology (CURE) study also observed the mean salt intake as 8.5 g/day, using the food frequency questionnaire (FFQ)32. The mean salt intake in women in an urban resettlement colony in Delhi was 7.6 g/day based on spot urine sodium measurements13. Some methods, such as dietary surveys, are time-consuming and have a low validity33,34. Even samples of dried urine obtained from spot collection samples are a clinically valid and convenient alternative to liquid urine to analyse sodium concentration under field conditions35. The salt intake was significantly higher in men, the rural population, the employed, overweight and obese. Similar findings on higher salt intake in males36,37, rural people30, and obese individuals38,39 were seen in other studies.

The salt added during food preparation constitutes the primary source of dietary salt in low- and middle-income such as India40. Studies conducted in high-income countries observed the average salt intake of 9–12 g daily; processed foods comprised the primary source41,42. However, with the socio-economic progress and changing lifestyle, India is witnessing a ‘nutrition transition’ wherein there is a decrease in the consumption of wholesome foods and an increasing reliance on packaged and processed food43,44.

Our study showed that the mean dietary salt intake to be high in the Indian population, which calls for planning and implementing control of dietary salt consumption measures. The results pave the way for the planning and implementing control of dietary salt consumption measures in the Indian population. A universal reduction in the dietary sodium consumption of at least 1.2 g per day would help achieve a 50% reduction in the proportion of persons who require anti-hypertensive treatment45. An awareness of the adverse health effects could substantially influence the willingness to curb excessive salt consumption. The findings on perceptions and practices related to salt intake could help design effective consumer education and awareness programmes. However, awareness generation alone may not always successfully bring about behavioural change46. A multisectoral and coordinated approach is essential to ensure actual limitations of salt intake. Intensive efforts should be leveraged through National Multi-sectoral Action Plan, which includes salt reduction as an active component. The Food Safety and Standard Authority of India (FSSAI) has played a significant role in generating awareness and encouraging community involvement through ‘The Salt Challenge, Every Pinch Counts’47. Instead of sodium chloride, low sodium salt substitutes have potassium chloride for reducing sodium content, although these are more expensive than regular salt. Framing of policies to scale up the availability and affordability of low sodium salt substitutes should be promoted to ascertain a reduction in the mean population salt intake. Other potential measures include mass awareness, training of food vendors and clear labelling of sodium content in packaged foods.

Strength and limitations

Our study was conducted in a nationally representative sample wherein dietary sodium intake was estimated from spot urine samples, a validated method used to assess dietary sodium intake. The population mean was calculated using sampling weights; thus, the study findings could be generalised at a population level and used to plan and implement dietary salt control measures. The information on awareness and behaviour of salt intake was self-reported and could be subjected to information bias. The study did not capture data on the actual dietary sources of salt in food items and condiments for any correlation analysis.