Abstract
Background: PAHO-WHO reports that sodium intake is currently high in the Caribbean. The objective was to estimate sodium (Na) and potassium (K) intakes by 72 h dietary recall and compare them with those obtained from 24 h urinary excretion in Dominican adults. Methods: A total of 69 adults (33 men) completed a 3-day dietary recall with emphasis on added salt and seasonings. The 24 h urine samples were analysed by indirect potentiometry using the membrane ion-selective electrode technique. The WHO-PAHO Questionnaire on Knowledge, Attitudes and Behaviour toward Dietary Salt and Health was completed. Results: Dietary Na intake ranged from 1.0 to 8.3 g. Median dietary and urinary Na concentrations were similar (2.7 and 2.5 mmol/d). Mean dietary Na and K concretertentrations were higher than those excreted in 24 h urine (133.0 ± 59.7 vs. 103.7 ± 44.5 mmol Na/d, p = 0.001; 69.0 ± 21.0 vs. 36 ± 16.3 mmol K/d, p < 0.001). The Na-to-K ratio was lower in dietary than in 24 h urine samples (2.0 ± 1.1 vs. 3.2 ± 1.6 mmol/d, p < 0.001). Urinary Na concentration was associated with sex (r = 0.280, p = 0.020) and obesity (r = 0.244, p = 0.043) and K with sex (r = 0.356, p = 0.003). Urinary Na-to-K was inversely related to age (r= -0.291, p = 0.015). Sex and obesity explained 11% of the variance in urinary Na concentration and sex only of the variance in urinary K concentration. The only significant correlation between dietary and urinary concentrations was that of K (r = 0.342, p = 0.004). This correlation matrix, controlled for overweight and sex, maintained the level of significance and was equal in almost 12% of the data. Conclusions: These data, which are the first data on Na and K intakes in Dominicans assessed by dietary assessment, showed a higher mean sodium intake (mean of dietary recall and urinary excretion data: 2.7 g Na, 6.8 g salt/day) and a lower K intake (2.06 g/day) than the WHO recommendations (<2.0 g Na, ≥3.5 g K). Potassium, but not sodium, intake from 72 h food recall and 24 h urinary excretion showed a correlation when controlling for sex and obesity, but not enough to consider them interchangeable.