Iodine deficiency has recently been found in UK young and pregnant women, which is of concern given the importance of adequate iodine intake in pregnancy for fetal brain development. The WHO recommends that iodine deficiency in a population should be corrected through salt iodisation but there is a lack of UK data on iodised-salt availability, a situation that the present study aimed to address.

Availability of iodised salt for household use was determined by a shelf survey in five supermarket chains in each of sixteen UK areas (in Southern England, Wales and Northern Ireland) encompassing a total of seventy-seven supermarkets. All branches of a sixth supermarket chain that had 2·3 % of the market share sold exclusively iodised salt. Weighted iodised-salt availability was calculated taking the market share of supermarkets into account.

The UK.

Not applicable.

Iodised salt was available in thirty-two of the seventy-seven supermarkets (41·6 %). After accounting for market share and including all six UK supermarket chains, the weighted availability of iodised salt was 21·5 %. The iodine concentration of the major UK brand of iodised salt is low, at 11·5 mg/kg.

In contrast to other countries, iodised household table salt is unlikely to contribute meaningful amounts to UK iodine intake as (i) availability is low, (ii) table salt is only a small percentage of total UK salt intake and (iii) UK public-health campaigns have encouraged reduced salt consumption. As iodine intake in the UK is dependent entirely on food choices, regular monitoring of iodine status is essential.

Sugar-sweetened beverage (SSB) consumption has declined steadily. This study uses the latest national data to examine trends in SSB consumption among children and adults by race and/or ethnicity and to document whether long-standing disparities in intake remain.

Trend analyses of demographic and dietary data measured by 24-h dietary recall from the National Health and Nutrition Examination Survey (NHANES).

Data from the 2003–2004 through 2017–2018 NHANES survey cycles were analysed in 2020.

The study sample included 21 156 children aged 2–19 years and 32 631 adults aged 20+ years.

From 2003–2004 to 2017–2018, the prevalence of drinking any amount of SSB on a given day declined significantly among all race and/or ethnicity groups for children (non-Hispanic (NH) White: 81·6 % to 72·7 %; NH Black: 83·2 % to 74·8 %, Hispanic: 86·9 % to 77·2 %) and most race and/or ethnicity groups for adults (NH White: 72·3 % to 65·3 %; Hispanic: 84·6 % to 77·8 %). Consumption declined at a higher rate among NH Black and Hispanic children aged 12–19 years compared with their NH White peers; among NH Black children aged 6–11 years, the rate of decline was lower. Despite significant declines in per capita SSB energy consumption from soda and fruit drinks, consumption of sweetened coffee/tea beverages increased among older children and nearly all adults and consumption of sweetened milk beverages increased among NH White and Hispanic children.

SSB consumption has declined steadily for children and adults of all race and/or ethnicity groups, but disparities persist, and overall intake remains high.

To investigate the association between the weight status of first–year female students (FYFS) and various weight management–related characteristics to identify possible components of a weight management programme for students.

Cross‐sectional study.

Female residences at a South African university.

A total of 360 FYFS.

Mean (± standard deviation (SD)) body mass index (BMI) of the FYFS was 21.8 ± 2.6 kg m−2, with 7.2% being underweight, 81.9% normal‐weight, 10.0% overweight and 0.8% obese. Underweight, normal–eight and overweight students differed with regard to their perception of their weight (P < 0.001), weight goals (P < 0.001) and previous weight–loss practices (P < 0.001). Mean ± SD score on the 26-item Eating Attitudes Test (EAT–26) was 8.5 ± 9.0 with 8.4% classified as high scorers. Mean ± SD score on the 34-item Body Shape Questionnaire (BSQ) was 87.7 ± 32.2, with 76.1% classified as low, 11.9% as medium and 11.9% as high scorers. The self-concept questionnaire indicated that 36.7% had a high, 43.9% a medium and 19.4% a low self-concept. Higher BMI correlated with a higher BSQ score (P < 0.001), a lower self-concept (P = 0.029) and a higher EAT-26 score (P < 0.001). Smoking was prevalent amongst 13.1% of students, and 51.2% used vitamin and/or mineral supplements. Students who quitted smoking had higher (P = 0.006) BMI (22.7 ± 2.9 kg m-2) than those who never smoked before (21.6 ± 2.5 kg m−2). Normal-weight students were more physically active than underweight or overweight students (P = 0.038).

The specific weight management-related needs of FYFS include information about supplement use, smoking, realistic weight goals, safe and sound weight-loss methods, weight cycling, body-shape perceptions, eating attitudes and behaviours, self-concept and physical activity. Interventions aimed at correcting these problems should target all students, regardless of their BMI.

To examine changes in breast-feeding take-up rates among young children in Scotland and to assess whether maternal education or occupation-based social class is a stronger and better predictor of breast-feeding take-up.

Binary logistic regression models were developed from the first sweep of the Growing Up in Scotland longitudinal survey, for the two cohorts of children.

A national representative survey for Scotland.

A baby cohort of 5012 singletons born over a 12-month period between June 2004 and May 2005, and a toddler cohort of 2732 singletons born over a 12-month period between June 2002 and May 2003.

Mothers from more privileged social classes and those with more educational qualifications resulted as more likely to breast-feed. However, maternal education was a better and more robust predictor of breast-feeding take-up compared with social class. There were no significant differences in breast-feeding take-up between the two cohorts and only minor differences between mothers aged 20–29 years and those who stated an intention to bottle-feed prior to birth.

The study suggests that the importance of maternal education in influencing breast-feeding has been somewhat overlooked in research based in more developed countries. The results indicate that, compared with occupation-related social class, maternal education is a more informative, accurate and useful lens through which to understand and explain patterns of breast-feeding take-up.

Large portion sizes in restaurants have been identified as a public health risk. The purpose of the present study was to determine whether customers in two different food-service operator segments (non-commercial worksite cafeteria and commercial upscale restaurant) would select reduced-portion menu items and the impact of selecting reduced-portion menu items on energy and nutrient intakes and plate waste.

Consumption and plate waste data were collected for 5 weeks before and 7 weeks after introduction of five reduced-size entrées in a worksite lunch cafeteria and for 3 weeks before and 4 weeks after introduction of five reduced-size dinner entrées in a restaurant setting. Full-size entrées were available throughout the entire study periods.

A worksite cafeteria and a commercial upscale restaurant in a large US Midwestern metropolitan area.

Adult worksite employees and restaurant patrons.

Reduced-size entrées accounted for 5·3–12·8 % and 18·8–31·3 % of total entrées selected in the worksite and restaurant settings, respectively. Food waste, energy intake and intakes of total fat, saturated fat, cholesterol, Na, fibre, Ca, K and Fe were significantly lower when both full- and reduced-size entrées were served in the worksite setting and in the restaurant setting compared with when only full-size entrées were served.

A relatively small proportion of reduced-size entrées were selected but still resulted in reductions in overall energy and nutrient intakes. These outcomes could serve as the foundation for future studies to determine strategies to enhance acceptance of reduced-portion menu items in restaurant settings.

To examine the contribution of the food service sector to the nutrient quality of the Irish diet, and to compare intakes at home, work and outside the home (‘ut’) and within the subgroups of the out location (pub, deli, takeaway)

Random sample of adults from the Republic of Ireland. Food intake data were collected using a 7-day food diary. Respondents recorded the location of every eating occasion determined by where the food was prepared rather than consumed

Intakes of energy, protein, fat and carbohydrate were significantly greater at home than at work or out (P <0.05). The intake of alcohol was significantly (P <0.001) greater out than at home or work. The percentage contribution of fat to energy was above the recommendations (33% of total energy and 35% of food energy) for both men and women at all locations, with the exception of the contribution of fat to total energy for men at the out location. Within the subgroups of the out location, the contribution of alcohol to total energy was greatest in pubs and the contribution of fat to both total and food energy was greatest in takeaways. Intakes of fibre and most micronutrients per 10 MJ of food energy were greater (P <0.05) at home than at work or out

Foods eaten outside the home contribute a disproportionately high level of fat intake and should be targeted in public health nutrition strategies

We examined the halo effect of a 2-year weight-loss diet trial, the Dietary Intervention Randomized Controlled Trial (DIRECT), on the weight and nutritional patterns of participants' spouses.

DIRECT participants in a research centre workplace were randomly assigned to one of three diets: Low-fat, Mediterranean or Low-carbohydrate. A sample of wives of the DIRECT participants, who attended support update meetings specific to their husband's diet during the first 6 months, were followed for 2 years.

South Israel.

Seventy-four women (mean age = 51 years, mean BMI =26·6 kg/m2).

Among the wives of husbands randomised to the Low-fat, Mediterranean and Low-carbohydrate diet, self-reported weight change was respectively −1·48 kg, −2·30 kg and −4·62 kg after 6 months, and +0·39 kg, −3·00 kg and −2·30 kg after 2 years. Weight loss among wives whose husbands were in the alternative diet groups combined (Mediterranean+Low-carbohydrate) was significantly greater than among wives whose husbands were in the Low-fat group after 6 months (P = 0·031) and 2 years (P = 0·034). Overweight wives experienced more weight loss. The weight change of couples was significantly correlated (r = 0·42, P < 0·001). Across all dietary groups, wives had significant improvement in their dietary patterns in all food groups according to their husbands' diets, mainly by a larger significant decrease in carbohydrate consumption in the Low-carbohydrate group (P = 0·013 compared to Low-fat). Six-month weight change among the seventy-four DIRECT participants whose wives took part in the group support sessions was −5·2 kg, compared to −3·5 kg among the 248 DIRECT participants whose wives did not take part in these sessions (P = 0·020).

Focusing on the couple as a unit could provide a cost-effective approach to weight-loss programmes.

To assess the relationship between education and the intake of a variety of individual foods, as well as groups of foods, for Australian men and women in different age groups.

Cross-sectional national survey of free-living men and women.

A sample of 2501 men and 2739 women aged 18 years and over who completed the National Nutrition Survey (NNS) 1995.

Information about the frequency of consumption of 88 food items was obtained using a food-frequency questionnaire in a nation-wide nutrition survey. Irregular and regular consumers of foods were identified according to whether they consumed individual foods less than or more than once per month. The relationship between single foods and an index of education (no post-school qualifications, vocational, university) was analysed via contingency table chi-square statistics for men and women. Food group variety scores were derived by assigning individual foods to conventional food group taxonomies, and then summing the dichotomised intake scores for individual foods within each food group. Two-way analyses of variance (education by age groups) were performed on food variety scores for men and women, separately.

While university-educated men and women consumed many individual foods more regularly than less-educated people, they were less likely to be regular consumers of several meat products. The relationship between education and food consumption was less apparent when individual food scores were aggregated into food group scores. University-educated men and women exhibited higher scores on total food group variety than the other educational groups.

Higher education is associated with the regular consumption of a wider variety of foods. Aggregation of individual food consumption indices into food variety scores may mask the apparent effects of educational background on food consumption.

To establish BMI percentile curves that describe the contemporary BMI distribution among Chinese children, and to compare their BMI percentile curves with those in two recently developed international references: the WHO and the US Centers for Disease Control and Prevention (US CDC) growth references.

A cross-sectional national survey.

Thirty provinces, municipalities and autonomous regions in China.

Nationally representative sample of 232 140 school students aged 7–18 years.

BMI percentile curves were established using the LMS method, and were compared with the percentiles of the WHO and the US CDC references. BMI distributions and growth patterns in Chinese children were dramatically different from those in the two international reference populations. Compared with the international reference populations, younger Chinese boys (7–12 years of age) had higher values of the percentiles above the median and lower values of the percentiles below the median, suggesting that they had larger proportions of extreme BMI values in both directions. Chinese girls and older Chinese boys (15–18 years of age) had substantially lower BMI percentiles than their counterparts in the reference populations, particularly those high percentiles among older age groups.

The present study described the unique patterns of BMI curves at the national level, and these curves are useful as a reference for comparing different regions and for monitoring changes over time in Chinese children. Higher proportions of children with extreme values in both directions indicate that China is currently facing both an increasing level of obesity and a high level of undernutrition, simultaneously.

Recommendations for fruit and vegetable consumption are largely unmet. Lower socio-economic status (SES), neighbourhood poverty and poor access to retail outlets selling healthy foods are thought to predict lower consumption. The objective of the present study was to assess the interrelationships between these risk factors as predictors of fruit and vegetable consumption.

Cross-sectional multilevel analyses of data on fruit and vegetable consumption, socio-demographic characteristics, neighbourhood poverty and access to healthy retail food outlets.

Survey data from the 2002 and 2004 New York City Community Health Survey, linked by residential zip code to neighbourhood data.

Adult survey respondents (n 15 634).

Overall 9·9 % of respondents reported eating ≥5 servings of fruits or vegetables in the day prior to the survey. The odds of eating ≥5 servings increased with higher income among women and with higher educational attainment among men and women. Compared with women having less than a high-school education, the OR was 1·12 (95 % CI 0·82, 1·55) for high-school graduates, 1·95 (95 % CI 1·43, 2·66) for those with some college education and 2·13 (95 % CI 1·56, 2·91) for college graduates. The association between education and fruit and vegetable consumption was significantly stronger for women living in lower- v. higher-poverty zip codes (P for interaction < 0·05). The density of healthy food outlets did not predict consumption of fruits or vegetables.

Higher SES is associated with higher consumption of produce, an association that, in women, is stronger for those residing in lower-poverty neighbourhoods.

To undertake a systematic review to determine similarities and differences in metrics and results between recently and/or currently used tools, protocols and methods for monitoring Australian healthy food prices and affordability.

Electronic databases of peer-reviewed literature and online grey literature were systematically searched using the PRISMA approach for articles and reports relating to healthy food and diet price assessment tools, protocols, methods and results that utilised retail pricing.

National, state, regional and local areas of Australia from 1995 to 2015.

Assessment tools, protocols and methods to measure the price of ‘healthy’ foods and diets.

The search identified fifty-nine discrete surveys of ‘healthy’ food pricing incorporating six major food pricing tools (those used in multiple areas and time periods) and five minor food pricing tools (those used in a single survey area or time period). Analysis demonstrated methodological differences regarding: included foods; reference households; use of availability and/or quality measures; household income sources; store sampling methods; data collection protocols; analysis methods; and results.

‘Healthy’ food price assessment methods used in Australia lack comparability across all metrics and most do not fully align with a ‘healthy’ diet as recommended by the current Australian Dietary Guidelines. None have been applied nationally. Assessment of the price, price differential and affordability of healthy (recommended) and current (unhealthy) diets would provide more robust and meaningful data to inform health and fiscal policy in Australia. The INFORMAS ‘optimal’ approach provides a potential framework for development of these methods.

To examine the prevalence of household food insecurity and compare dietary intake by food security status in a representative Korean population.

Cross-sectional. Food security status of households was classified using an eighteen-item food security questionnaire. The nutrition survey comprised questions on dietary habits, a 24 h dietary recall and a semi-quantitative FFQ.

The 2012 Korea National Health and Nutrition Examination Survey.

A total of 3007 households completed the food security questionnaire. Family members within each household aged ≥1 year (n 7118) participated in the nutrition survey.

Results from the 2012 survey indicated that 88·7 % of Korean households showed food security. The remaining 11·3 % (9·3 % for food insecurity without hunger and 2·0 % for food insecurity with hunger) were in food-insecure households. The prevalence of household food insecurity was 13·2 % in households with children and 10·3 % in households without children. Mean daily intakes of energy, fat and carbohydrates were not significantly different between food-secure and food-insecure adults. In contrast, mean daily intakes of protein, crude fibre, vitamins and minerals as well as weekly consumption frequencies of vegetables, seaweeds, fruits, fruit juice, nuts, and milk and milk products were significantly lower in food-insecure adults compared with food-secure adults.

The study demonstrated that food insecurity is associated with reduced intakes of healthy foods and nutrients essential for health and growth in a representative Korean population.

The objective of the present study was to examine the relationship of dietary fried fish consumption and risk of cardiovascular events and all-cause mortality.

Prospective cohort study among participants of the REasons for Geographic And Racial Differences in Stroke (REGARDS) study who resided in the USA.

The primary outcome measures included the hazard ratios (HR) of incident CVD including first incident fatal or non-fatal ischaemic stroke or myocardial infarction and all-cause mortality, based on cumulative average fish consumption ascertained at baseline.

Participants (n 16 479) were enrolled between 2003 and 2007, completed the self-administered Block98 FFQ and were free of CVD at baseline.

There were 700 cardiovascular events over a mean follow-up of 5·1 years. After adjustment for sociodemographic variables, health behaviours and other CVD risk factors, participants eating ≥2 servings fried fish/week (v. <1 serving/month) were at a significantly increased risk of cardiovascular events (HR=1·63; 95 % CI 1·11, 2·40). Intake of non-fried fish was not associated with risk of incident CVD. There was no association found with dietary fried or non-fried fish intake and cardiovascular or all-cause mortality.

Fried fish intake of two or more servings per week is associated with an increased risk of cardiovascular events. Given the increased intake of fried fish in the stroke belt and among African Americans, these data suggest that dietary fried fish intake may contribute to geographic and racial disparities in CVD.

To estimate the proportion of ‘healthy’ snack food and beverage choices available to an Australian consumer.

A survey of product Nutrition Information Panels (NIP) and product labels on snack foods and beverages offered for sale. Data on nutrient content were compared with criteria from different nutrient profile systems to estimate the proportion of items conforming to a ‘healthy’ choice.

A large supermarket in metropolitan Melbourne, Australia.

A consumer could choose from 1070 different snack foods and 863 different drinks. Flavour variety was more common in snacks (maximum thirteen per product) while variation in container size was more common for drinks (up to ten per product). Recommended serving size for snacks varied greatly (18–100 g) while the serving size for drinks frequently did not correspond to the size of the container. Depending on the nutrient profile system selected, only 9–22 % of snack foods presented for sale could be deemed ‘nutritious’ by multiple criteria. Similarly, only 14–27 % of beverages met ‘healthy’ criteria.

As one factor to help reduce the obesogenic environment, the supply balance needs to be shifted in favour of ‘healthier’ snack foods and beverages, e.g. by reformulation of many products by the food industry and their presentation in smaller, standardised portion-size packaging.

To assess the association between total alcohol intake, specific alcoholic beverages and sleep quality in a community-based cohort.

A cross-sectional study.

The Kailuan community, China.

Included were 11 905 participants who were free of a history of CVD, cancer, Parkinson’s disease, dementia and head injury in or prior to 2012. Alcohol consumption (amount and frequency intake) and alcoholic beverage type were collected in 2006 (baseline) and 2012. Participants were grouped into non-, light- (women: 0–0·4 serving/d; men: 0–0·9 serving/d), moderate- (women: 0·5–1·0 serving/d; men: 1·0–2·0 servings/d) and heavy- (women: >1·0 servings/d; men: >2·0 servings/d) drinkers. Overall sleep quality was measured in 2012 and included four sleep parameters (insomnia, daytime sleepiness, sleep duration, snoring/obstructive sleep apnoea).

We observed a dose–response association between higher alcohol consumption in 2006 and worse sleep quality in 2012 (Ptrend < 0·001), after adjusting for age, sex, socio-economic status, smoking status, physical activity, obesity, plasma lipid profiles, diabetes and hypertension. A similar association was observed when alcohol consumption in 2012 was used as exposure. Alcohol was associated with higher odds of having short sleep duration (adjusted OR for heavy- v. non-drinkers = 1·31; 95 % CI: 1·09, 1·57) and snoring (adjusted OR for heavy- v. non-drinkers: 1·38; 95 % CI: 1·22, 1·57). Consumption of hard liquor, but not beer or wine, was significantly associated with poor sleep quality.

Higher alcohol consumption was associated with poorer sleep quality and higher odds of having snoring and short sleep duration.

To describe dietary changes in men participating in an obesity intervention as part of the Self-Help, Exercise and Diet using Information Technology (SHED-IT) study.

An assessor-blinded randomized controlled trial comparing Internet (n 34) v. information-only groups (n 31) with 6-month follow-up. Dietary intake assessed by FFQ, reporting usual consumption of seventy-four foods and six alcoholic beverages using a 10-point frequency scale. A single portion size (PSF) factor was calculated based on photographs to indicate usual serving sizes.

The campus community of the University of Newcastle, New South Wales, Australia.

Sixty-five overweight/obese men (43 % students, 42 % non-academic general staff, 15 % academic staff; mean age 35·9 (sd 11·1) years, mean BMI 30·6 (sd 2·8) kg/m2).

The average PSF decreased significantly over time (χ2 = 20·9, df = 5, P < 0·001) with no differences between groups. While both groups reduced mean daily energy intake (GLM χ2 = 34·5, df = 3, P < 0·001), there was a trend towards a greater reduction in the Internet group (GLM χ2 = 3·3, P = 0·07). Both groups reduced percentage of energy from fat (P < 0·05), saturated fat (P < 0·001) and energy-dense/nutrient-poor items (P < 0·05), with no change in dietary fibre or alcohol (P > 0·05).

Although men reported some positive dietary changes during weight loss, they did not increase vegetable intakes nor decrease alcohol consumption, while saturated fat, fibre and Na intakes still exceeded national targets. Future interventions for men should promote specific food-based guidelines to target improvements in their diet-related risk factor profile for chronic diseases.

Nitrate and nitrite are probable human carcinogens when ingested under conditions that increase the formation of N-nitroso compounds. There have been limited efforts to develop US databases of dietary nitrate and nitrite for standard FFQ. Here we describe the development of a dietary nitrate and nitrite database and its calibration.

We analysed data from a calibration study of 1942 members of the NIH–AARP (NIH–AARP, National Institutes of Health–AARP) Diet and Health Study who reported all foods and beverages consumed on the preceding day in two non-consecutive 24 h dietary recalls (24HR) and completed an FFQ. Based on a literature review, we developed a database of nitrate and nitrite contents for foods reported on these 24HR and for food category line items on the FFQ. We calculated daily nitrate and nitrite intakes for both instruments, and used a measurement error model to compute correlation coefficients and attenuation factors for the FFQ-based intake estimates using 24HR-based values as reference data.

FFQ-based median nitrate intake was 68·9 and 74·1 mg/d, and nitrite intake was 1·3 and 1·0 mg/d, in men and women, respectively. These values were similar to 24HR-based intake estimates. Energy-adjusted correlation coefficients between FFQ- and 24HR-based values for men and women respectively were 0·59 and 0·57 for nitrate and 0·59 and 0·58 for nitrite; energy-adjusted attenuation factors were 0·59 and 0·57 for nitrate and 0·47 and 0·38 for nitrite.

The performance of the FFQ in assessing dietary nitrate and nitrite intakes is comparable to that for many other macro- and micronutrients.

To evaluate the association of vegetable and fruit intake with several demographic, anthropometric, lifestyle and dietary factors.

Cross-sectional analysis. Information on habitual diet was collected by means of the diet history method. Association of vegetable and fruit intakes with other factors was assessed separately by means of multiple lineal regression and the cumulative odds model.

39 622 healthy subjects aged 29–69 years from the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort in Spain.

Fruit intake increased with age, education and physical activity and decreased with intake of saturated fatty acids and cholesterol; smoking and alcohol consumption were also negatively associated with fruit, with a very low consumption for current smokers and heavy drinkers. Vegetable intake increased with education and physical activity and with intake of unsaturated fatty acids, mainly mono-unsaturated. Former smokers consumed more vegetables than never or current smokers and non-consumers of alcohol ate less vegetables than consumers, among whom no differences were observed. All these estimates were adjusted by energy and body mass index (BMI).

When assessing the association of fruit and vegetables with chronic diseases it is important to take into account confounding factors. Furthermore, it would be useful to study dietary patterns including several interrelated factors.

To identify child hunger and examine its association with family factors, receipt of benefits, housing conditions and social support among recently arrived refugee families with young children.

Structured and semi-structured questionnaire administered to a service-based, purposive sample of caregivers.

East London, United Kingdom.

Thirty households with children <5 years old, resident in the UK for <2 years.

All households sampled were food-insecure, and 60% of index children were experiencing hunger as defined on the Radimer/Cornell scale. Child hunger was significantly associated with recent arrival, marginally significantly associated with receipt of fewer benefits and younger parenthood, and not associated with maternal education or self-efficacy score, household size or composition, or measures of social support.

A community-based, participatory approach for rapid assessment of the prevalence, extent and causes of child hunger among newly arrived asylum seekers recently arrived in Britain is feasible, and preliminary results suggest a programmatic need for a broader, population-based assessment of food insecurity in this rapidly growing population group.

To assess whether nutritional deficiency increases susceptibility to arsenic-related health effects.

Assessment of nutrition was based on a 24 h recall method of all dietary constituents.

Epidemiological cross-sectional study was conducted in an arsenic-endemic area of West Bengal with groundwater arsenic contamination.

The study was composed of two groups – Group 1 (cases, n 108) exhibiting skin lesions and Group 2 (exposed controls, n 100) not exhibiting skin lesions – age- and sex-matched and having similar arsenic exposure through drinking water and arsenic levels in urine and hair.

Both groups belonged to low socio-economic strata (Group 1 significantly poorer, P < 0·01) and had low BMI (prevalence of BMI < 18·5 kg/m2: in 38 % in Group 1 and 27 % in Group 2). Energy intake was below the Recommended Daily Allowance (set by the Indian Council of Medical Research) in males and females in both groups. Increased risk of arsenical skin lesions was found for those in the lowest quintile of protein intake (v. highest quintile: OR = 4·60, 95 % CI 1·36, 15·50 in males; OR = 5·62, 95 % CI 1·19, 34·57 in females). Significantly lower intakes of energy, protein, thiamin, niacin, Mg, Zn and choline were observed in both males and females of Group 1 compared with Group 2. Significantly lower intakes of carbohydrate, riboflavin, niacin and Cu were also observed in female cases with skin lesions compared with non-cases.

Deficiencies of Zn, Mg and Cu, in addition to protein, B vitamins and choline, are found to be associated with arsenical skin lesions in West Bengal.