Estimation of mean population salt intakes using spot urine samples and associations with body mass index, hypertension, raised blood sugar and hypercholesterolemia: Findings from STEPS Survey 2019, Nepal

Background High dietary salt intake is recognized as a risk factor for several non-communicable diseases (NCDs), in particular cardiovascular diseases (CVDs), including heart attack and stroke. Accurate measurement of population level salt intake is essential for setting targeted goals and plans for salt reduction strategies. We used a spot urine sample to estimate the mean population salt intake in Nepal and evaluated the association of salt intake with excess weight, hypertension, raised blood sugar and hypercholesterolemia, and a number of socio-demographic characteristics. Methods A population-based cross-sectional study was carried out from February to May 2019 using a WHO STEPwise approach to surveillance. Spot urine was collected from 4361 participants aged 15–69 years for the analysis of salt intake. We then used the INTERSALT equation to calculate population salt intake. Student’s ‘t’ test, one-way ANOVA and multivariable linear regression were used to assess the association between salt intake and a number of factors. Statistical significance was accepted at P < .05. Results The average (±SD) age of participants was 40 (14.1) years. Mean salt intake, derived from spot urine samples, was estimated to be 9.1g/d. A total of 70.8% of the population consumed more than the WHO’s recommended amount of 5g salt per day, with almost one third of the population (29%) consuming more than 10g of salt per day. Higher salt intake was significantly associated with male gender (β for male = 0.98g; 95%CI:0.87,1.1) and younger age groups (β25–39 years = 0.08; 95%CI:-0.08,0.23) and higher BMI (β = 0.19; 95%CI:0.18,0.21). Participants who were hypertensive and had raised blood cholesterol consumed less salt than people who had normal blood pressure and cholesterol levels (P<0.001). Conclusions Salt consumption in Nepal is high, with a total of 70.8% of the population having a mean salt intake >5g/d, well above the World Health Organization recommendation. High salt intake was found to be associated with sex, age group, education, province, BMI, and raised cholesterol level of participants These findings build a strong case for action to reduce salt consumption in Nepal in order to achieve the global target of 30% reduction in population salt intake by 2025.


Meghnath Dhimal
Background: High dietary salt intake is recognized as a risk factor for several non-26 communicable diseases (NCDs), in particular cardiovascular diseases (CVDs), including heart 27 attack and stroke. Accurate measurement of salt intake is essential for setting realistic goals and 28 plans for salt reduction strategies. We used spot urine sample for the first time to estimate the 29 mean population salt intake in Nepal. We also evaluate the association of salt intake with BMI, 30 Hypertension, raised Blood Sugar and Blood Lipids and their relation with socio-demographic 31

characteristics. 32
Methods: A population based cross sectional study was carried out from February to May 2019 33 using a WHO STEP-wise approach to surveillance. 4361 (67.6%) spot urine was collected from 34 the men and women aged 15-69 for the analysis of salt intake. INTERSALT equation was used 35 to calculate population salt intake. Student's't' test, ANOVA and multivariate linear regression 36 regressions was used to assess the association between salt intake with explanatory factors. 37 Statistical significance was accepted at P<.05. 38 Results: The average (±SD) age of participants was 40 (14.1) years. Mean salt intake was 39 estimated to be 9.1g/d derived from spot urine samples. A total of 69.4% of the population 40 consumed more than the WHO's recommended amount of 5 g salt per day, with almost half of 41 the population 48.9% consuming more than 10 g of salt per day. Higher salt intake was 42 significantly associated with male gender (β for male = 0.98g; 95% CI: 0.87, 1.1) and younger 43 age groups (β25-39 years = 0.08; 95% CI: -0.08, 0.23) and higher BMI (β = 0.19; 95% CI: 0.18, 44 0.21). Participants who were hypertensive, and had high cholesterol ate less salt than people who 45 had normal blood pressure and cholesterol level (P==0.000).

54
Sodium in the salt is the principal cation in extracellular fluid in the body, and is an essential 55 nutrient necessary for normal cell function and for neurotransmission [1,2,3]. Salt is essential 56 nutrient if consumed in limit. High dietary salt intake is associated with high blood pressure 57 which is recognized as a risk factor for non-communicable diseases (NCDs), in particular 58 cardiovascular diseases (CVDs), including heart attack and stroke [4,5,6]. Conclusive scientific 59 evidence also found the association of excessive consumption of sodium with osteoporosis, 60 cataract, kidney stones, and diabetes [4,6]. Globally, populations are consuming excessive 61 amounts of salt, with the worldwide estimated mean salt intake being almost 9-12 g/day [7], in 62 many countries and even higher intakes are found in Asia, which is associated with negative 63 effects on health [8]. Step 1: This step comprised a Questionnaire to gather demographic and behavioral characteristics of the study population; face to face interview was carried out to fill the 113 questionnaire. In Step 2: Anthropometric measurement: Blood pressure, height, weight, hip, 114 waist circumference were measured. Weight was measured with a portable digital weighing scale 115 (Seca, Germany). Waist and hip circumference was measured using a constant tension tape 116 (Seca, Germany). Blood pressure was measured by using digital, automated blood pressure 117 monitor (OMRON digital device) with a universal size cuff.
Step 3: Biochemical measurements 118 were undertaken to determine the proportion of the study population with diabetes, raised blood 119 glucose and abnormal lipid level. Blood glucose and total cholesterol was measured through dry 120 chemistry using CardioCheck PA Analyser as recommended and supported by WHO. 121 Concentrations of glucose, total cholesterol were measured in capillary whole blood. Fasting 122 samples was taken to measure raised blood glucose. Participants were instructed to fast overnight 123 for 12 hours at the time of household visit for Step 1 and 2. 124 125 Estimation of 24-hour salt intake based on sport urine testing 126 For the estimation of the mean population salt intake, STEPs survey utilized spot urine sample as 127 a proxy to 24-hour urine samples. Spot urine collection was done to identify the level of Sodium 128 (Na), potassium (K) and creatinine. 129 For urine collection, urine container with QR code was provided to participants to collect spot 131 urine. Participants self-collected the urine samples at home before fasting for blood sample 132 collection the next day during their scheduled appointment for biochemical measurements. 133 The collected urine sample was stored in dark place in normal room temperature until they were 134 transported to the lab. Laboratory setup was done in every province headquarters and nearly 135 located places. Determination of Na and K in urine is carried with Ion-selective Electrodes (ISE) 136 in an automated Analyzer (Beckman Coulter, CA, USA) and creatinine was estimated in urine by 137 the use of semi-automated biochemistry analyser (Nova Biomedical Cooperation, Waltham, MA, 138 USA). The unit of measurements for Na and K was mmol/L, while creatinine was mg/dl. 139 Participants who were pregnant, fasting before collecting the urine sample, having contaminated 140 urine sample with blood were excluded at the time of analysis. Participants whose height was 141 less than 100 cm or above 270 cm; weight was less than 20kg or above 350 kg were also 142 excluded.  Among the male respondents, more than one fourth (29.8%) of population were assessed as having hypertension. Similarly, among the female respondents about 19.7% respondents had 214 high blood pressure (i.e., SPB ≥140 mm Hg and/or DBP ≥90 mm Hg). More female participants 215 (11.2 %) reported having previously been diagnosed with hypertension on treatment compared to 216 (7.0%) male participants. Diabetes mellitus (DM) was prevalent among male than female (6.3% 217 compared to 5.3) in contrast female respondents had high cholesterol level than male (13.9% 218 compared to 7.7%). More female participants reported of ever had a heart attack or chest pain 219 from heart disease or stroke (1.4 % compared to 0.8%) in male participants.  Mean salt intake was estimated to be 9.1g/d derived from spot urine samples " Table 2". Males 227 had significantly higher salt intake than females (9.6 g/d compared to 8.7 g/d). A total of 69.4% 228 of the population consumed more than the WHO's recommended amount of 5 g salt per day, 229 with almost half of the population 48.9% consuming more than 10 g of salt per day. The 230 sodium/potassium ratio for the population was 3.4 (SE, 0.2). 231 232

Education
None/less than primary 2152 8.9 (8.9-9.0) 0.000 Male participants consumed on average 1g more salt than female. Young adult (25-39years) 247 participant ate more salt 0.08g than older (55-69 years) age group participants 248 (P==0.000). Although there was no significant difference observed, more than secondary 249 education participants ate less salt than participants who had primary and secondary education. 250 Salt intake decreases as wealth index of the participant's increases Compared to Province 1, 251 participants living in Bagmati, Gandaki Province ate less salt compared to participants living in Karnali and Sudurpashchim Province however, there was no significant association was 253 observed among participants residing in Province 2 and Lumbini Province. Salt intake increased 254 by 0.2g as BMI increased (0.19) (P==0.000). Participants who were hypertensive, and had high 255 cholesterol ate less salt than people who had normal blood pressure and cholesterol level 256 (P==0.000). Compared to non-diabetic participants diabetic participants ate less salt although 257 this difference was nonsignificant " Table:

population. 270
We found that mean 24-h salt intake was 9.6g/day for men and 8.7 g/day for women, with mean 271 salt intake 9.1g/day among adult Nepalese population, higher than WHO recommended amount 272 of 5g/day and majority of population (69.4%) in our context consumed more salt than this 273 recommendation. Interestingly, our study reported 31% of the population consumed over 274 10g/day salt which is double than the WHO recommended value. The average mean salt 275 consumption level in our study is comparable to small scale studies conducted in Nepal [8] Our finding reported the sodium/potassium ratio for the studied population was 3.4 (SE, 0.2), 294 which is consistent with the finding in previous study done by Samoa et al [45]. Reducing the 295 Na/K ratio is essential for preventing hypertension and cardiovascular disease; however there is 296 no generally accepted recommended guideline for the Na/K ratio [46,47]. Future studies are 297 required to establishing the Na/K ratio for providing information to individuals regarding the risk 298 of hypertension and cardiovascular disease [47]. The finding of this study suggested that higher 299 value of salt intake in male than female which is consistent with other studies 300 [8,32,35,36,45,48,49]. Most of the male population in Nepal is engaged in outdoor activity and 301 most often consume prepared or ready to eat foods which might leads to greater sodium 302 consumption among male than female. This study shows young adult (25-39 years) ate more salt 303 as compared to age group between (15-24 years), however, salt consumption is decline in 304 middle (40-54 years) and older (55-69 years) age group people which is not surprising as similar 305 finding were observed in the study conducted in Nepal [36]. Nepalese economic structure has 306 changed shifting away from agricultural food supply system towards modern processing food supply system. The trade liberalization has made processed foods easily available at 308 supermarkets and fast food outlets [50]. Our findings also align with the current study conducted 309 in Lifestyle Practices and Obesity in Nepalese Youth which shows majority (75.78%) of 310 respondents consumes fast-food [51]. Similarly, people with higher grade levels were 311 significantly highly likely to be knowledgeable about risk factors of non-communicable disease 312 [52] , which also reflect in this study as salt intake is low in population who had education more 313 than secondary level. Risk factors of non-communicable diseases increased with increasing 314 wealth [53,54], however; in contrast to these findings our study shows salt consumption was 315 decline in those people who were included in fourth and highest wealth index. This may be due 316 to fact that wealthier population may be aware about risk factor of non-communicable disease 317 and may change their lifestyle. Our study showed people who reside in Karnali and 318 Sudhurpashchim province ate more salt than other province. Salt intake is increased as BMI 319 increased in line with another study which shows that salt intake was higher in overweight and 320 obese individuals, this may lead to susceptible to non-communicable disease in later life among 321 those population[33,36,45,55,56]. The correlation of high salt intake with obesity is well known, 322 but the biological mechanisms behind this correlation are not well understood yet. High sodium 323 intake has been suggested as an indirect cause of obesity through increased thirst following 324 consumption of highly salted foods causing an increased intake of sugar-sweetened soft drinks 325 Unlike other studies conducted elsewhere [56,59,60,61], we didn't find any significant 327 association between increase salt intake and hypertension. It may be due to the fact that 328 hypertensive individuals are aware about the amount of salt intake as they are advised by the 329 treating physician and dietician about the amount of salt intake in their diet. A possible reason 330 for this is that patient with raised blood pressure and under medication advice to restrict sodium 331 intake or phenotype of salt sensitivity is heterogeneous, influenced from genetic to 332 environmental factors with multiple mechanisms that potentially link high salt intake to increases 333 in blood pressure [62]. However, our finding reported salt intake was less among hypertensive 334 population compared to non-hypertensive. Sodium reduction substantially lowered blood 335 pressure, even among those with starting systolic blood pressure levels as low as 120 mm Hg 336 hypertension with ageing, as the association between sodium intake and blood pressure is greater 340 at older age [64]. High-salt intake is a major risk factor for developing hypertension in type 2 341 diabetes mellitus, but its effects on glucose homeostasis are controversial [65] . Similarly, we did 342 not find the association of salt intake in diabetes population though there was decrease in salt 343 intake in population who are diabetic than non-diabetic. This study finds association between salt 344 intake and blood lipids, however there is no established conclusive evidence regarding the 345 relationship between sodium intake and blood lipid level [61,66,67,68]. Study has shown mixed 346 evidence on reducing sodium intake may had or no have significant adverse effect on blood 347 lipids and association between sodium intake and all-cause mortality, incident cardiovascular This study has several strength, first of it analyzed first time the salt intake in community level in 354 Nepalese population, and is not limited to any particular region, caste, sex, and age. Second, to 355 our best knowledge, this is the first study in Nepalese population which demonstrated the 356 association of salt intake with different characteristic however, this study has also few major 357 limitation; being a cross sectional study, it cannot attribute the causality from the association of 358 estimated salt intake and BMI and blood pressure. Second, the use of spot urine sample for the 359 estimation of salt intake instead of its gold standard method 24-h urine sample. Therefore, its 360 validation study in small Nepalese population is recommended. However, many large 361 epidemiological studies have already adopted the estimation of salt intake using spot urine 362 sample instead of 24-h urine sample because of the ease of urine sample collection and 363 participant enrollment. Third, this study has included hypertensive individual but their detailed 364 information on antihypertensive drug was not inclusively collected via a questionnaire. It is 365 postulated that some antihypertensive drugs with natriuretic properties could be a confounder, 366 but the other antihypertensive drug could not be the confounder. To be sure about the 367 confounding factor, a correlation study with exposure namely salt intake in this study is required. in Nepalese population. The estimation of salt intake using spot urine samples is likely to discriminate that salt intakes are well above the WHO recommendations and may provide a as a 377 benchmark to assess the impact of salt reduction efforts in Nepal [19]. 378 379

380
Our study finding suggested that higher estimated daily salt intake, approximately double than