C.A. Farleigh

Attitudes and personality related to salt intake

A predominantly middle-class female sample of British adults completed questionnaires designed to estimate salt intake and to assess attitudes towards adding table salt to food, along with Cattells 16PF personality inventory. The attitudes questions were based on the approach suggested by Fishbein and Ajzen, and proved to be good predictors of table salt use. The persons own feeling about table salt use was a better predictor than perceived pressure from other people. Beliefs about adding table salt were closely related to the persons attitude, but only beliefs about taste (not beliefs about health consequences) predicted usage. Beliefs about pressure from doctors and nutritionists/dietitians, but not family, were also related to usage. There was lower usage and more negative attitudes in respondents under 35 years old. Two of the second-order personality factors from the 16PF correlated with salt intake, but appeared to act through attitudes: tough poise (i.e. given to rapid action with insufficient thought) correlated positively with estimated table and cooking salt use; anxiety correlated positively with cooking use, with salt consumed in foods and with estimated total intake. The results suggest that any attempt to reduce salt intake might best be directed at satisfying taste, and not exclusively at augmenting anxiety about the health consequences of salt intake.

Preference and sensitivity to salt taste as determinants of salt-intake.

Total salt-intake was estimated in 33 normal subjects using urinary sodium excretion over 7 days, and table and cooking salt use were measured over the same period. Sensitivity to salt taste was measured for 31 of these subjects using a seven-category rating scale for intensity of saltiness with five concentrations of salt in water, bread and mashed potato. Preferences were assessed using a nine-category hedonic rating scale with the bread and mashed potato to determine the maximally-preferred concentration. No differences were found in the taste measures between subjects divided into high and low intake in terms of total salt-intake or total intake divided by body weight. However, differences were found in sensitivity and preferences between groups divided into high and low table salt use. This may be because the major proportion of total salt-intake arose from salt present in foods and this would not be so open to voluntary control as table salt use.

THE RELATIONSHIP BETWEEN SALT INTAKE AND PREFERENCES FOR DIFFERENT SALT LEVELS IN SOUP

Two experiments are reported with subjects divided according to questionnaire responses into low and high total sodium intake, and low and high table salt use. The stimuli were tomato soup with varying salt concentrations. In Experiment 1 these were rated for salt intensity, pleasantness and intensity relative to ideal. In Experiment 2 the stimuli were presented in restricted concentration ranges or using a method designed to minimise range bias; relative-to-ideal intensity rating was again used. Subjects with a high total intake were found to prefer higher concentrations of salt assessed both with the hedonic and relative-to-ideal intensity ratings, though not with restricted stimulus ranges. Subjects with high table salt use were found to prefer higher concentrations only when assessed using the hedonic ratings. Sensitivity to salt taste was not found to differ between the groups.

Limited compensation by table salt for reduced salt within a meal

Sixteen subjects, all of whom had said in a preliminary questionnaire that they normally added table salt to foods, were fed standard meals in the laboratory over 10 days. The meals were identical, except that on 5 days the meal had no added salt (containing 0.46 g sodium chloride) or had salt added to a level of 5.09 g. They were allowed free access to salt pots with the meals and used an average of 1.40 g table salt with the unsalted meal and 0.36 g with the salted meal, thus compensating for 22% of the difference in salt content of the meal. There was no difference in water consumption between the two types of meal. Nutrient intake from the rest of the diet did not differ between periods with high and low salt meals. The failure to compensate more fully for reduced salt in the foods can be attributed to the greater availability of table salt for perception; less table salt than salt incorporated in the foods is therefore required. Reduction of salt concentrations in purchased foods would be unlikely to be fully replaced by the consumer adding table salt.

Effects of haemodialysis on taste and thirst

Fifteen patients undergoing haemodialysis tasted soup varying in salt concentration and apple puree varying in sucrose concentration, immediately before and after dialysis. Matched controls tasted the same foods with a similar interval between tastings. The samples were rated for intensity, and on a relative-to-ideal scale. For the salt, the slopes of the functions plotted against log (concentration) were higher after dialysis than before, whilst the most preferred concentration was lower. There were no effects found for the sweetness ratings or for the controls. Likewise there were no overall differences in the ratings between the patients and controls. Thirst was found to increase on dialysis, and there was a trend of this being higher for the patients than for the controls. The reduction in preferences for salt by dialysis would make compliance with a reduced salt diet easier, but the increase in thirst would make compliance with reduced fluid intake more difficult.

Estimation of salt intake by questionnaire

A questionnaire designed to assess salt intake was tested. This involved principal components analysis of responses from 155 subjects to find the underlying components related to the questions. Validation of the questionnaire used 7-day urinary sodium excretion for 33 subjects, along with table and cooking salt usage over 7-days. This gave comparatively good prediction for total intake (r = 0.66) and table salt use (r = 0.70) but not for cooking salt use (r = 0.17). Test-retest reliability for the estimate of total intake was r = 0.75. This method would therefore offer a useful estimate of total intake, especially where group means are required rather than accurate estimates of individual intakes.

The relationship between intensity, hedonic and relative-to-ideal ratings

Abstract Forty five subjects rated soup samples varying in salt concentration on three types of ratings, intensity, hedonic and relative-to-ideal. Estimates of the most preferred concentration for individual subjects from the hedonic and relative-to-ideal ratings correlated at r = 0.76 ( P P

Effects of Stimulus Context on Preference Judgements for Salt

An experiment is reported which confirms earlier findings that the range of stimulus concentration affects the estimate of ideal concentration. The method proposed by Booth et al for reducing range bias has been found to give satisfactory results.

Changes in Salt Taste in Dialysis and Their Relationship to Blood Constituents

Taste sensitivity for sodium chloride in solution and sensitivity and preference for sodium chloride in bread were measured in 14 male patients undergoing hemodialysis. There was an increase in sensitivity to salt taste following dialysis only when assessed using the bread samples. Patients on a lower sodium diet were more sensitive to salt taste in bread. Although the taste changes were not related to the decreases in serum sodium or to copper, greater increases in sensitivity were related to greater increases in serum zinc during dialysis.

The effect of manipulation of salt pot hole size on table salt use

Abstract Daily use of table salt was measured for 10 weeks in a staff canteen serving approximately 2400 meals. During the study the size of the holes in the salt pots was manipulated, and the corresponding amount of table salt used was measured. Staff also completed a questionnaire on salt use, which was used to calculate the percentage of staff using the salt pots. Results indicate that, despite a 10 day exposure to each new hole size, individuals used significantly more table salt when the salt pots with the largest holes were supplied. However, when the ‘shaking’ time was estimated by relating the amount used to the flow rate, the salt pots with the largest holes were shaken for a shorter time. This finding suggests that the adding of table salt to food is subject to both sensory and habitual control.