The Scottish Health Survey 2011: Volume 1 - Adults

CHAPTER 5 DIET

Rachel Whalley

SUMMARY

• In 2011, the mean number of portions of fruit and vegetables consumed per day by adults was 3.2. Mean daily consumption was significantly higher for women (3.3) than for men (3.1).
• One in five (22%) adults met the recommended daily intake of five or more portions of fruit and vegetables (20% of men and 23% of women). The proportion of adults meeting the recommendation has not changed significantly over time.
• Adults aged 16-24 consumed the fewest portions per day (2.6 portions) and were also the age group least likely to consume five or more portions a day (15%). 17% of 16-24 year olds did not consume any fruit and vegetables in the 24 hours prior to interview.
• Men had higher mean urinary sodium (119.4mmol/l) and potassium levels (65.3 mmol/l) than women in 2008-2011 (mean levels for women were 95.0 mmol/l and 58.1 mmol/l respectively.
• Urinary sodium levels decreased by age for both men and women. The mean level for those aged 16-44 was 122.0 mmol/l compared with a mean of 87.2 mmol/l for those aged 65 and over. Urinary potassium levels were also highest in the youngest age group (64.5 mmol/l).
• Between 2003 and 2008-2011 there was a significant decline in both the mean sodium level for adults (from 116.1 mmol/l to 106.9 mmol/) and the mean creatinine level (from 12.2 mmol/l to 10.5 mmol/l. Urinary potassium levels remained unchanged over this same period.
• Over a quarter (27%) of women and a fifth (20%) of men took some type of vitamin or mineral supplement in 2008-2011. Consumption was lowest among those aged 16-24 (13%) and highest among those aged 65-74 (36%).
• The proportion of adults taking a dietary vitamin or mineral supplement was slightly lower in 2008-2011 than in 2003 (24% and 26% respectively).
• There was a significant association between supplement consumption and SIMD in 2008-2011. Consumption tended to decline as deprivation level increased with 28% of those living in the least deprived quintile taking a vitamin or mineral supplement compared with 17% of those living in the most deprived quintile.

5.1 INTRODUCTION

This chapter covers three areas related to dietary habits: self-reported fruit and vegetable consumption, direct measurement of sodium and potassium levels via urine samples vitamin supplement use. The dietary supplement and urine sample results have not been reported since the 2003 Scottish Health Survey (SHeS) report.^1,2

Scotland's unhealthy diet is widely cited as a factor in its poor health record. In particular, low consumption of fruit and vegetables is a risk factor for cardiovascular disease, cancer, hypertension, type 2 diabetes and obesity, while excess salt consumption has been linked to hypertension.

The World Health Organisation (WHO) recommends adults eat at least five varied portions - where a portion is defined as 80g - of fruit and vegetables a day. Detailed information about fruit and vegetable consumption (designed to measure adherence to the '5 a day' recommendation) was first collected in the 2003 survey, and has been included every year since 2008.

Significant efforts have been taken in recent years to encourage the public to consume less salt, and industry to use less salt in food production (the majority of dietary sodium intake is derived from processed foods rather than its direct addition to food at the table). Sodium is a vital constituent of the body and thus an essential nutrient.¹ However, the relationship between salt intake and health, in particular cardiovascular disease, is well-established. Scientific evidence suggests that a high salt intake contributes to the development of high blood pressure and the Scientific Advisory Committee on Nutrition (SACN) concluded that reducing the average salt intake of the population is likely to decrease the burden of high blood pressure and improve public health.^1,2,3 The Scottish Diet Action Plan,⁴ the Scottish Dietary Targets,⁵ and the 2008 action plan to combat obesity - Healthy Eating, Active Living⁶ - all share a common commitment to reduce population-level salt intake to no more than 6g per day (2.4g or 100mmo/l of sodium).⁷ The two most recent estimates for adults aged 19-64 in Scotland, based on follow-up studies of SHeS participants, showed that levels of salt intake were similar in 2006 (9.0g) and 2009 (8.8g), and were in excess of the recommended 6g.^8,9 SHeS has collected urine samples to assess levels of salt intake (urinary sodium), potassium and creatinine since 2003.

While most people should be able to obtain all the nutrients required to maintain good health from a balanced diet, NHS Scotland recommends that additional vitamin and mineral supplements are necessary for adults in certain circumstances.¹⁰ These are: folic acid for women trying to conceive and in the first 12 weeks of pregnancy; vitamin D for all pregnant and breastfeeding women, people aged 65 and over, people with darker skins, and those who may not be exposed to much sunlight (e.g. housebound people). In addition, people with restricted diets and certain medical conditions may be advised by a clinician to take additional supplements. Following concern about possible vitamin D deficiency in the population, the UK's four Chief Medical Officers reissued their guidance about vitamin D supplementation to remind health professionals about the recommendations outlined above.¹¹ Information about overall dietary supplement use was collected in the nurse visits in the 2003 and 2008-2011 surveys.

The equivalent dietary chapters in the 2008, 2009 and 2010 SHeS reports,^12,13,14 provided overviews of the broader dietary policy context from the mid 1990s onwards, some of which has been mentioned above. They outlined a number of actions taken by the Government and NHS Scotland to improve diets in Scotland, including initiatives designed to encourage more fruit and vegetable consumption, in line with the recommendation to eat at least five portions of fruit and vegetables a day, and as already mentioned, to reduce salt consumption. These included:

• The Scottish Diet Action Plan,⁴ which outlined the Scottish Dietary Targets.⁵
• The White Paper Towards a Healthier Scotland.¹⁵
• The Scottish Executive's Improving Health in Scotland - the Challenge paper.¹⁶
• The Hungry for Success initiative.¹⁷
• A framework for implementing the Diet Action Plan: Eating for health meeting the challenge.¹⁸
• The Scottish Government's Better Health, Better Care Action Plan.¹⁹
• Healthy Eating, Active Living: An action plan to improve diet, increase physical activity and tackle obesity (2008-2011).⁶
• The Scottish Government's Obesity Route Map,²⁰ and associated Obesity Route Map Action Plan.²¹

Between 2008 and 2011 only a sub-sample of participants were invited to have an additional nurse interview. For this reason the analysis of urinary sodium and potassium, and of vitamin / mineral supplement use, presented here is based on either two or four years of nurse data combined. From 2012 the survey is no longer including a nurse visit and instead a sub-sample of adults will be asked to complete a new biological module, conducted by specially trained interviewers. Spot urine samples are part of this new module so the trends over time will be maintained. Questions about vitamin supplement have also been retained (as part of the main interview).

This chapter updates the trends in fruit and vegetable consumption among adults since 2003. Urinary sodium and potassium levels in 2003 are compared with the more recent figures for 2008-2011 combined. Consumption of vitamin or mineral supplements in 2003 is also compared with the 2008-2011 period, and the most recent figures are also presented by the Scottish Index of Multiple Deprivation (SIMD).

5.2 METHODOLOGY

5.2.1 Measures of eating habits

Two different modules of questions were used to assess eating habits. One of these assessed fruit and vegetable consumption, and was designed with the aim of providing sufficient detail to monitor the '5-a-day' policy effectively. This module was asked of all adults and children aged 2 and over every year between 2008 and 2011. The second module was asked of all children every year, and a sub-sample of adults in 2008 and 2010. It used a modified version of the Dietary Instrument of Nutrition Education (DINE) questionnaire developed by the Imperial Cancer Research Fund's General Practice Research Group to assess participants' usual intake of a wide range of nutrients, including protein, starch, fat and fibre.²² This chapter only reports the findings from the fruit and vegetable module for adults.

To determine the total number of portions that had been consumed in the 24 hours preceding the interview, the fruit and vegetable module asked about the following food types: vegetables (fresh, frozen or canned); salads; pulses; vegetables in composites (e.g. vegetable chilli); fruit (fresh, frozen or canned); dried fruit; and fruit in composites (e.g. apple pie). A portion was defined as the conventional 80g of a fruit or vegetable. As 80g is difficult to visualise, a 'portion' was described using more everyday terms, such as tablespoons, cereal bowls and slices. Examples were given in the questionnaire to aid the recall process, for instance, tablespoons of vegetables, cereal bowls full of salad, pieces of medium sized fruit (e.g. apples) or handfuls of small fruits (e.g. raspberries). In spite of this, there may be some variation between participants' interpretation of 'a portion'. These everyday measures were converted back to 80g portions prior to analysis. The following table shows the definitions of the portion sizes used for each food item included in the survey:

Food item	Portion size
Vegetables (fresh, frozen or canned)	3 tablespoons
Pulses (dried)	3 tablespoons
Salad	1 cereal bowlful
Vegetables in composites, such as vegetable chilli	3 tablespoons
Very large fruit, such as melon	1 average slice
Large fruit, such as grapefruit	Half a fruit
Medium fruit, such as apples	1 fruit
Small fruit, such as plum	2 fruits
Very small fruit, such as blackberries	2 average handfuls
Dried fruit	1 tablespoon
Fruit in composites, such as stewed fruit in apple pie	3 tablespoons
Frozen fruit/canned fruit	3 tablespoons
Fruit juice	1 small glass (150 ml)

Since the '5-a-day' policy stresses both volume and variety, the number of portions of fruit juice, pulses and dried fruit was capped so that no more than one portion could contribute to the total number of portions consumed. Interviewers recorded full or half portions, but nothing smaller.

5.2.2 Urinary sodium and potassium

Dietary salt intake is assessed by measuring sodium excretion in urine. The studies on which the estimates in Section 5.1 were based on involved analyses of urine samples collected over a 24-hour period.^8,9 24 hour urine collection is accepted as being the most reliable method for assessing salt intake in the population.¹ A less burdensome measure, based on a spot sample collected at one point in time, has been included in the SHeS nurse interview since 2003. While the absolute level of sodium measured will differ between the spot and 24-hour samples, previous validation studies showed that spot urine samples could assess trends over time, and differentiate between population sub-groups, in the same way as 24-hour samples.^23,24 As spot samples are less burdensome to collect than 24 hour samples the number of people asked to provide them is usually higher than in the 24-hour collection studies allowing for more detailed sub-group analyses to be conducted.

A spot urine sample was collected in all nurse interviews conducted between 2008 and 2011, and in a sub-sample of nurse interviews in the 2003 survey, to determine dietary sodium (Na). As discussed in the introduction there is a target to reduce population-level salt intake to no more than 6g per day (2.4g or 100mmo/l of sodium).²⁵

To aid the analysis of dietary sodium, spot urine samples were also assessed for potassium and creatinine. Potassium is important for digestion, metabolism and muscle tissue regulation and abnormally high levels of potassium are indicative of hyperkalaemia. Similarly abnormally low levels can be problematic. The usual range for adults with a regular diet is 25-125 (mmol/L). Creatinine (Cre), a product of creatine, was included because while large day-to-day variations occur in excretion of Na, K and water, Cre excretion is relatively constant from day-to-day (coefficient of variation 11%).²⁶ Therefore the ratio of Na and K excretion to creatinine excretion is normally used in the literature to correct for variability in urine dilution (random urine specimen). The association between Na/Cre ratios and blood pressure has been reported in several studies. Na/Cre and K/Cre ratios vary from day-to-day, however these ratios are less sensitive to incompleteness of urine specimens than the individual Na, K or Cre excretion. See Volume 3 (Technical Report) for further details of the measurement protocols for the urine samples.

5.2.3 Vitamin and/or mineral supplement consumption

In 2003 and 2008-2011, the nurse visit included the following question design to measure self-administered supplement use:

At present, are you taking any vitamins, fish oils, iron supplements, calcium, other minerals or anything else to supplement your diet or improve your health, other than those prescribed by your doctor?

Participants were presented with a list of possible supplement types on a card and asked to say which they used. The options were: vitamins, fish oils, iron supplements, calcium, other minerals, other supplements. The tables in this chapter report the total proportion who said they were taking supplements as well as the proportions taking each of the specific supplement-types asked about.

5.3 FRUIT AND VEGETABLE CONSUMPTION

5.3.1 Trends in adult consumption of fruit and vegetables since 2003

Information on the quantity of fruit and vegetables men and women aged 16 and over had consumed in the 24 hours prior to the interview is presented for 2003 onwards in Table 5.1. The table includes the mean and median number of portions consumed, as well as the proportions who met the daily recommended consumption of five or more portions.

In 2011, the mean number of portions of fruit and vegetables consumed by adults aged 16 and over was 3.2 - the same as in 2010. The separate figures for men and women were also identical in 2010 and 2011 (3.1 mean portions for men and 3.3 for women). This suggests that the small, but significant, increase from 3.1 portions in 2003 to 3.3 portions in 2008 did not constitute a meaningful sustained trend. It also highlights the problems of comparing single figures in a time series, rather than assessing underlying trends.

In line with the trend for mean consumption, the proportion of adults consuming the recommended five or more portions of fruit or vegetables a day did not change significantly over time. In both 2010 and 2011, 22% met the recommendation. This was preceded by 23% in 2009, 22% in 2008 and 21% of adults meeting the recommendation in 2003, which suggests an overall picture of trendless fluctuation. When examined separately, the recent consumption figures for men and women confirm this unchanging picture. The proportion of men meeting the recommended daily intake has remained noticeably static across recent years (22% in 2009 and 20% in all others, including 2011). The 2010 and 2011 results for women were the same (23%), which confirms that what appeared to be an upward trend in consumption between 2003 and 2009 (from 22% to 25%) has not been sustained. Table 5.1

5.3.2 Portions of fruit and vegetables consumed by age and sex, 2011

More detailed figures for the quantity of fruit and vegetables consumed in the 24 hours prior to the interview for adults aged 16 and over in 2011 by age and sex are also presented in Table 5.1. In addition, figures 5A and 5B show the summary measures of five or more portions, no portions and the mean number, by age for men and women separately.

As noted above, adults consumed on average 3.2 portions of fruit and vegetables per day in 2011. While small, the difference between the mean number of portions consumed by men (3.1) and women (3.3) was statistically significant. Consumption varied with age, with the youngest adults (aged 16-24) consuming the lowest number of portions (2.6), and the figures for adults aged 25 and over ranging between 3.3 and 3.4 portions per day. As shown in Figures 5A and 5B, this pattern by age was evident for both men and women.

The proportion of adults who met the recommended daily intake of five or more portions of fruit and vegetables per day also varied significantly by gender. While 22% of all adults met the recommendation, women were more likely to do so than men (23% compared with 20%). The overall association between age and meeting the recommendation was not significant, but at 11%, the proportion of men aged 16-24 who ate five or more portions a day was significantly lower than for all other age groups (21%-23%). Figure 5A, Figure 5B, Table 5.1

5.4 URINARY SODIUM, POTASSIUM AND CREATININE

5.4.1 Trends in urinary sodium, potassium and creatinine since 2003

Table 5.2 shows the levels of sodium (Na), potassium (K), creatinine (Cre) and the Na/Cre ratio and K/Cre ratio from spot urine samples in 2003 and 2008-2011 combined. Mean and median levels, as well as levels for the 5^th, 10^th, 90^th and 95^th percentile are presented by age and sex.

Between 2003 and 2008-2011 there was a statistically significant decline in the mean urinary sodium level for adults aged 16 and over from 116.1mmol/l to 106.9mmol/l. This decline brings the level closer to the population-level target of no more than 2.4g or 100mmol/l of sodium (6g of salt) per day. This reduction was reflected in the fact that levels of urinary sodium at the upper end of the distribution (the 95^th percentile) were lower in 2008-2011 than in 2003 (213 mmol/l compared with 222 mmol/l). For a combination of reasons including differing methodologies, time periods and sample sizes, these results are not comparable with the results of the Scottish Salt Studies referenced in the introduction.^8,9 It is also important to note that it is not uncommon for sodium concentrations in spot urine samples to be lower than the levels found in the 24 hour urine samples.

Creatinine levels followed a similar trend to urinary sodium, with a decrease in the mean level from 12.2 mmol/l to 10.5 mmol/l. In contrast, mean urinary potassium levels were broadly similar in 2003 (62.5 mmol/l) and 2008-2011 (61.6 mmol/l) with mean levels that fell within the range. The usual range for adults with a regular diet is 25-125 (mmol/l)

In line with these findings, the ratios of sodium to creatinine (Na/Cre) and potassium to creatinine (K/Cre) both increased over time. In 2003 Na/Cre was 12.2; in 2008-2011 it was 12.9. Similarly, the ratio for K/Cre in 2003 was 5.9 and increased to 6.8 in 2008-2011. Table 5.2

5.4.2 Urinary sodium, potassium and creatinine by age and sex, 2008-2011 combined

More detailed figures on urinary sodium and potassium levels by age and sex in 2008-2011 combined are presented in Table 5.2. The mean urinary sodium level in 2008-2011 was 106.9mmol/l. This is in excess of the population-level target of no more than 6g of salt per day (2.4g or 100mmol/l of sodium) for the adult population. In line with findings in 2003, men had significantly higher mean levels of sodium than women (119.4mmol/l compared with 95.0mmol/l). That the mean level was higher for men than for women is not unexpected. Once caloric intake adjustments are made, the target of 6g/day (2.4g or 100mmol/l of sodium) for adults represents 7g/day (2.7g or 115mmol sodium) for men and 5g/day (2.0g or 85mmol/l sodium) for women. Mean levels for men and women in 2008-2011 were both higher than these adjusted figures. Sodium levels varied significantly with age for both men and women with levels decreasing as age increased (from 122.0mmol/l in adults aged 16-44 to 87.2 mmol/l in those aged 65 and over). This varying pattern by age was noted in 2003.

Differences in urinary potassium levels were in keeping with those seen for sodium, with higher mean levels among men (65.3 mmol/l) than women (58.1 mmol/l). The usual range for adults with a regular diet is considered to be 25-125 (mmol/l). Levels varied by age and were higher among younger adults aged 16-44 (64.5mmol/l) compared with those aged 65 and over (54.4mmol/l). The pattern by age differed for men and women, with levels declining successively across the three age groups among women but only declining between the two oldest age groups among men.

The ratios of sodium to creatinine (Na/Cre) and potassium to creatinine (K/Cre) both followed the same patterns as those for mean urinary sodium and potassium levels, with lower ratio levels for men than for women, and ratio levels increasing with age for both men and women. These patterns were consistent with the 2003 results. Table 5.2

5.5 CONSUMPTION OF VITAMIN AND MINERAL SUPPLEMENTS

5.5.1 Trends in vitamin and mineral supplement consumption since 2003

The proportion of men and women consuming vitamin and mineral supplements in 2003 and 2008-2011 combined is presented in Table 5.3. In addition to showing the proportion consuming any supplement, information on consumption of specific types of supplement such as fish oils, calcium and iron is also shown.

The proportion of adults who reported taking any dietary supplements was slightly lower in 2008-2011 than in 2003 and (24% and 26%, respectively). However, this overall figure masks the fact that among some sub-groups, most notably women aged 45-64, there was a much higher than average decrease in supplement use (of eight to ten percentage points). Table 5.3

5.5.2 Vitamin and mineral supplement consumption by age and sex, 2008-2011 combined

Although the questionnaire cannot be used to establish which types of vitamin or mineral people take, the results showed that only a minority of those aged 65 and over took vitamins or minerals regularly, indicating low adherence to the recommendation for vitamin D.

More detailed figures on vitamin and mineral supplement consumption by age and sex for 2008-2011 combined are presented in Table 5.3.

In 2008-2011, a greater proportion of women (27%) than men (20%) took any type of supplement. This pattern was true for all but the youngest age group, so was not, therefore, caused by a higher prevalence of supplement use among women of child bearing age (for whom supplements are recommended before and during pregnancy). The pattern was also apparent for consumption of specific types of supplement, and was most pronounced for vitamins or minerals (14% of women compared with 10% of men) and other supplements (8% and 5% respectively).

Supplement consumption in 2008-2011 was lowest among those aged 16-24 (13%) and increased with age to a peak of 36% of those aged 65-74, before declining somewhat among those aged 75 and over (33%). This pattern was largely similar for both men and women, although men's supplement use flattened out among the three oldest age groups, while women's use continued to increase until age 75 at which point it declined. Table 5.3

5.5.3 Vitamin and mineral supplement consumption (age-standardised) by Scottish Index of Multiple Deprivation (SIMD), 2008-2011 combined

Table 5.4 presents vitamin and mineral supplement use by the SIMD. Two measures of SIMD are being used throughout this report. The first, which uses quintiles, enables comparisons to be drawn between the most and least deprived 20% of areas and the intermediate quintiles. The second contrasts the most deprived 15% of areas with the rest of Scotland (described in the tables as the "85% least deprived areas"). A description of SIMD is available in the Glossary at the end of this Volume). To ensure that the comparisons presented in this section are not confounded by the different age profiles of the SIMD sub-groups, the data have been age-standardised (age-standardisation is also described in more detail in the Glossary). Only the age-standardised data are presented in the tables in this section.

There was a significant association between supplement consumption and SIMD in 2008-2011. Supplement consumption was similar among among those living in the 4th and 5th least deprived areas (27%-28%), dropped to 23% in the next two quintiles, before falling to a low of 17% among those living in the most deprived quintile. The gradient of the decline in consumption was a little steeper for women than men. A third of women (33%) living in the least deprived areas took supplements compared with a fifth (20%) of those in the most deprived areas. The equivalent figures for men were 22% and 14%, respectively.

Comparing consumption among those living in the 15% most deprived areas of Scotland with those living elsewhere confirms the significant association between consumption and deprivation. 17% of adults in the 15% most deprived areas consumed a supplement compared with 25% in the rest of Scotland and this difference was evident for both men and women. Table 5.4