The Scottish Health Survey 2011: Volume 1 - Adults

CHAPTER 7 OBESITY

Linsay Gray and Alastair H Leyland

SUMMARY

• In 2011, over a quarter (27.7%) of adults aged 16 and over were obese (27.7% of men and 27.6% of women). Just under two-thirds (64.3%) were overweight or obese. Men were significantly more likely than women to be overweight or obese (69.2% compared with 59.6%).
• The mean Body Mass Index (BMI) in 2011 was 27.6 kg/m² for men and 27.5 kg/m² for women.
• Between 1995 and 2011, the proportion of adults aged 16-64 who were overweight or obese (BMI of 25 kg/m² and over) increased from 52.4% to 62.2%. Over this same period the prevalence of obesity (BMI of 30 kg/m² and over) among this age group also increased from 17.2% to 26.5%. The greatest increases were seen between 1995 and 2008 with figures remaining broadly stable since then.
• There was also an increase in mean BMI among adults aged 16-64 between 1995 and 2001 (from 25.8 kg/m² to 27.3 kg/m²). Again, the greatest increase occurred between 1995 and 2008 and has been largely stable since then.
• Obesity prevalence increased significantly with age in 2011, from 13.4% in those aged 16-24 to a peak of 35.4% in those aged 65-74. 16-24 year olds were least likely to be overweight including obese (36.0%) while those aged 65-74 were most likely to be (77.5%).
• In the 2010/2011 period, the mean waist circumference was 96.3cm for men and 89.0cm for women. Women were significantly more likely than men to have a raised waist circumference (49.1% compared with 31.7%).
• Based on a combination of their BMI and waist circumference measurements, women were more likely than men to be classified as being at high (or greater) risk of conditions like type 2 diabetes, hypertension and CVD (45.4% compared with 34.4% of men).
• Among men, the proportion at high (or greater) risk of such conditions increased with age up until age 55-64 at which point it levelled out. For women the proportion at high risk also increased with age but up until age 65-74 before dipping for those aged 75 and over.
• 15.7% of men were overweight according to their BMI but when the combined measure of BMI and waist circumference was used they were classified as being at no increased risk of obesity related diseases. The comparable figure for women was just 4.1%.
• There was a significant association between disease risk and both socio-economic classification and household income with clearer patterns observed for women than for men. Women living in semi-routine and routine households were the most likely to be classified as at a high (or greater) risk of obesity related disease whereas those in professional and managerial households were least likely to be (52.1% compared with 41.0%).
• Men living in the least deprived SIMD quintile were least likely to have health risks (49.1% had no increased risk, compared with 44.7%-46.6% of those living elsewhere. For women, the proportion at no increased risk decreased in line with deprivation (from 45.3% in the least deprived quintile to 29.8% in the most deprived).
• Age, economic status and physical activity levels were all independently significantly associated with being at high risk of disease for both men and women. For men, education level, marital status and self-assessed health status were also significant factors. For women, SIMD, parental NS-SEC, smoking status and presence of a long-standing illness were independently associated with being at high risk of disease.

7.1 INTRODUCTION

Obesity has a major impact on quality of life and health, increasing risk of type 2 diabetes, hypertension, cardiovascular disease, osteoarthritis and cancer.¹ Scotland has one of the worst obesity records amongst developed countries. The estimated cost to the NHS in Scotland of obesity and related illnesses in 2007/8 was in excess of £175 million.² With these economic and health costs, tackling obesity is a key priority for the public health sector in Scotland.

The introductions to the obesity chapters in the 2008,³ 2009⁴ and 2010⁵ Scottish Health Survey (SHeS) Reports provided a detailed overview of the recent policy context in Scotland. These included:

• The Scottish Government's Healthy Eating, Active Living: An action plan to improve diet, increase physical activity and tackle obesity.⁶
• The Keep Well initiative.⁷
• The Scottish Government's Route Map for tackling obesity and the associated Obesity Route Map Action Plan, published in 2011.⁸ SHeS is the measurement tool for seven of the Route Map's indicators, including the following long-term goal: the majority of Scotland's adult population in normal weight throughout life.⁹
• The Scottish Intercollegiate Guidelines Network (SIGN) national clinical guideline on obesity management.¹⁰

In addition, a number of policy actions targeted specifically at improving diets (described in Chapter 5) and physical activity levels (described in Chapter 6) are also relevant in the context of tackling obesity. Furthermore, as outlined in the chapter on child obesity in Volume 2, much of the effort to tackle unhealthy weight in the population is targeted at children, reflecting evidence that many children who are overweight or obese continue to be so in adulthood. For example, there are National Performance Framework National Indicators around healthy birthweight¹¹ and child healthy weight.¹²

This chapter focuses on body mass index (BMI) and waist circumference, derived from the direct measurements of height and weight taken in the main interview, and the waist measurements taken as part of the nurse visit. Time trends in BMI and waist circumference over the 1995-2011 period are examined by age and sex. Previous reports have also included data on waist/hip ratio. However, due to space constraints, concerns about the usefulness of this ratio as an indicator of obesity, and the fact that hip circumference is not being measured from 2012 onwards, this chapter only reports waist circumference results.

Between 2008 and 2011 only a sub-sample of participants was invited to have an additional nurse visit. For this reason the analysis of waist circumference 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. Waist circumference is part of this new module. A validation study has been conducted to assess the impact on the time series data of the change in methodology for measuring waist circumference.¹³ Future SHeS reports will discuss the implications in full.

The obesity chapter in the 2009 SHeS report included, for the first time, some analysis of disease risk using a measure recommended by the World Health Organisation, and endorsed in Scotland by SIGN, that takes into account both BMI and waist circumference.⁴ This chapter takes advantage of the larger sample provided by the 2008-2011 combined data to explore this further and presents disease risk by socio-economic classification, household income and the Scottish Index of Multiple Deprivation (SIMD).

7.2 METHODS AND DEFINITIONS OF MEASUREMENT

Full details of the protocols for carrying out the measurements are contained in Volume 3 of this report and are briefly summarised here.

7.2.1 Height

Height was measured using a portable stadiometer with a sliding head plate, a base plate and three connecting rods marked with a metric measuring scale. Participants were asked to remove shoes. One measurement was taken, with the participant stretching to the maximum height and the head positioned in the Frankfort plane.¹⁴ The reading was recorded to the nearest millimetre.

7.2.2 Weight

Weight was measured using Soehnle and Tanita electronic scales with a digital display. Participants were asked to remove shoes and any bulky clothing. A single measurement was recorded to the nearest 100g. Participants aged under 2 years, or who were pregnant, or chairbound, or unsteady on their feet were not weighed. Participants who weighed more than 130 kg were asked for their estimated weights because the scales are inaccurate above this level. These estimated weights were included in the analysis.

In the analysis of height and weight, data from those who were considered by the interviewer to have unreliable measurements, for example those who had excessive clothing on, were excluded from the analysis.

7.2.3 Body Mass Index (BMI)

The Body Mass Index (BMI), defined as weight (kg)/height (m²), is a widely accepted measure that allows for differences in weight due to height. It has been used in each SHeS report to date. However, BMI has some limitations.^15,16 It does not distinguish between mass due to body fat and mass due to muscular physique. It also does not take account of the distribution of fat.

BMI was calculated for all those participants for whom a valid height and weight measurement was recorded.

BMI classification

Adult participants were classified into the following BMI groups:¹⁷

BMI (kg/m2)	Description
Less than 18.5	Underweight
18.5 to less than 25	Normal
25 to less than 30	Overweight
30 to less than 40	Obese, excluding morbidly obese
40+	Morbidly obese

Other cut off points are also used in analyses of obesity, for example the World Health Organisation (WHO) cites evidence that chronic disease is an increasing risk in populations when BMI exceeds 21,¹⁸ while mortality rates do not necessarily correlate neatly with the categories presented here.¹⁹ However, meaningful comparisons of prevalence estimates between countries require agreed thresholds and these categories correspond with the WHO's recommended definitions for underweight, normal, overweight and obese (though they use three sub-classifications of obesity rather than the two presented here).²⁰ The tables by age and sex report both mean BMI and prevalence of the five categories outlined above. Although obesity has the greatest ill-health and mortality consequences, overweight is also a major public health concern, not least because overweight people are at high risk of becoming obese, while underweight also has negative health consequences. The trend tables present three measures: the proportion who is either overweight or obese (BMI of 25 kg/m² or more), the proportion who are obese (BMI of 30 kg/m² or more), and the proportion morbidly obese (BMI of 40 kg/m² or more). The latter group are at particularly high risk of morbidity and mortality.²¹

7.2.4 Waist measurements

Waist and hip measurements were conducted as part of the nurse interview. As noted in the introduction, only waist measurements are reported here.²² Waist was defined as the midpoint between the lower rib and the upper margin of the iliac crest. It was measured using a tape with an insertion buckle at one end. Each measurement was taken twice, using the same tape, and was recorded to the nearest even millimetre. Those participants whose two waist measurements differed by more than 3 cm had a third measurement taken.

For waist measurements, all those who reported that they had a colostomy or ileostomy, or were chairbound or pregnant, were excluded from the measurement. All those with measurements considered unreliable by the nurse, for example due to excessive clothing or movement, were excluded from the analysis.

Raised waist circumference

It has been postulated that waist circumference (WC) may be a better measure than BMI to identify those with a health risk from being overweight. The definition of raised WC used is in accordance with the definition of abdominal obesity used by the National Institutes of Health (USA) ATP (Adult Treatment Panel) III.²³ A raised WC has been taken to be more than 102 cm in men and more than 88 cm in women. These levels identify people at risk of metabolic syndrome, a disorder characterised by increased risk of developing diabetes and cardiovascular disease. Abdominal obesity is reported as more highly correlated with metabolic risk factors (high levels of triglycerides, low HDL-cholesterol) than elevated BMI.²³

7.2.5 WHO combined classification of disease risk

As noted in the introduction, the SIGN guideline on obesity¹⁰ cites the WHO's recommendation that an individual's risk of conditions such as type 2 diabetes and CVD is better estimated using a combination of both BMI and waist circumference (WC). The table below sets out the classification categories SIGN suggest. SIGN also note that increased WC can be a marker for disease even among people of normal weight. The analysis presented in this chapter classified people with normal weight and very high WC as at increased risk of disease. This chapter uses the BMI data collected in the main interview in combination with the waist measures collected by the nurse to estimate the proportion of the Scottish population who fall into each of the risk categories. This combined classification designates those with a raised WC as 'very high' WC, while those towards the upper end of the 'not raised' WC range are designated 'high' WC. As the table below indicates, the health risk is similar for adults with very high WC and class I obesity and for adults with high WC and class II obesity.

Type 2 diabetes, hypertension and CVD risk relative to normal weight and waist circumference

Classification	BMI (kg/m2)	Class	'High' WC Men WC 94-102cm Women WC 80-88cm	'Very high' WC Men WC >102cm Women WC >88cm
Normal weight	18.5 - <25		-	-
Overweight	25 - <30		Increased	High
Obese
Mild	30 - <35	I	High	Very high
Moderate	35 - <40	II	Very high	Very high
Extreme	40+	III	Extremely high	Extremely high

Source: based on Table 3, p11, in SIGN 115.¹⁰

7.3 RESPONSE TO ANTHROPOMETRIC MEASUREMENTS, BY AGE AND SEX

Response to height, weight, BMI, and waist and hip among adults (for 2008-2011 combined) is shown in Table 7.1. In previous reports the figures for single years have been presented, however since many of the data in this chapter are based on the 2008-2011 samples combined, the response figures are based on combined data also. A valid height measurement was obtained for 87% of men and 86% of women in this period. Response generally declined with age with the lowest levels among those aged 75 and over (74% of men and 70% of women aged 75 and over compared with 90% of those aged 16-24).

Valid weight measurements were provided by 86% of men and 84% of women. As with height, the proportions of men and women providing valid weight measurements were lowest for the oldest age group (75% of men and 72% of women). Proportions with known values for both height and weight, and thus derived BMI were similar to those for weight alone (85% of men and 83% of women), and followed similar patterns by age. Valid waist and hip measurements were obtained for almost all men (99%) and women (98%) who had a nurse visit; again response was slightly lower for those aged 75 and over (97% of men and 94% of women). Table 7.1

7.4 TRENDS IN THE PREVALENCE OF OVERWEIGHT AND OBESITY SINCE 1995

This section presents figures for the prevalence of overweight including obese (BMI 25 kg/m² or more), obesity (BMI 30 kg/m² or more), morbid obesity (BMI 40 kg/m² or more) and mean BMI by age for each survey year to date. Figures are presented for all adults and for men and women separately. Changes to the sample composition in the earlier survey years mean trends since 1995 are based on all adults aged 16-64, while trends for all adults aged 16 and over are presented for 2003 onwards. Adults' BMI in 2011 is discussed in more detail in Section 7.5.

7.4.1 Obesity and morbid obesity

As shown in Table 7.2 and Figure 7A, prevalence of obesity (BMI 30 kg/m² or more) among adults aged 16-64 in Scotland has risen significantly over the last sixteen years. Between 1995 and 2011 there was around a ten percentage point increase in the proportion of adults aged 16-64 that were obese (from 17.2% to 26.5%). As the more detailed discussion below illustrates, most of this increase occurred between 1995 and 2008, with the more recent figures showing some evidence of stability.

The increase in obesity over time followed a similar pattern for both men and women. For men aged 16-64, prevalence increased from 15.9% in 1995 to 22.0% in 2003 and then again to 24.9% in 2008. Between 2008 and 2011 it was fairly stable, ranging from 24.9% to 26.7%. The greatest increase for women also occurred between 1995 and 2003 (17.3% to 23.8%), with the figures since 2008 again, remaining fairly stable (ranging from 26.4% to 28.1%). While obesity prevalence in 2011 was significantly higher than in 1995-2003 it was not significantly different to levels in the 2008-2010 period.

The trend in obesity for all adults aged 16 and over since 2003 was similar to that discussed above for 16-64 year olds. A notable exception is that the increase in obesity for all adults was largely accounted for by rising levels among men with no significant increase among women. In 2003, 22.4% of men aged 16 and over were obese compared with 26.0%-27.7% from 2008 onwards. In contrast, the most recent figure for women (27.6%) was only a little higher than in 2003 (26.0%).

As noted in previous SHeS reports, morbid obesity prevalence (BMI of 40 kg/m² or more) is very low: just 2.9% of adults in 2011. However, this has also increased over time, from 1.2% of 16-64 year olds in 1995 to 2.7% in 2003, with levels fluctuating between 2.5% and 3.0% since 2008. The pattern for all adults aged 16 and over since 2003 was similar with prevalence fluctuating between 2.2% and 2.9%. Figure 7A, Table 7.2

7.4.2 Overweight and obesity

There has also been an increase over time in the proportion of 16-64 year olds that were overweight including obese (BMI 25 kg/m² or more) (from 52.4% in 1995, to 62.2% in 2011). As with the patterns in obesity discussed above, there was a large increase between 1995 and 2003, with the more recent figures being broadly stable (ranging from 62.2% to 63.3% in the 2008-2011 period). Prevalence of overweight including obesity has fluctuated more over this period for women than for men - after rising from 47.2% in 1995 to 57.3% in 2003, prevalence was then a little higher between 2008 and 2010 (58.4%-60.3%), but fell again in 2011 to 57.1%. This may well be trendless fluctuation, or the drop in 2011 could be the start of a decline in overweight including obesity prevalence among women aged 16-64; the 2012 and 2013 figures will help to answer this. In contrast, in recent years the proportion of overweight or obese men aged 16-64 has remained stable.

The pattern for all adults aged 16 and over since 2003 was similar to that for 16-64 year olds. For men, there was a small increase in overweight including obese prevalence between 2003 and 2008 (from 65.4% to 68.5%) followed by relative stability. In contrast, the proportion of women who were overweight or obese fluctuated, with no obvious pattern, between 59.6% and 62.4% and the 2003 and 2011 figures were very similar. Figure 7A, Table 7.2

7.4.3 Mean BMI

Mean BMI for adults aged 16-64 increased from 25.8 kg/m² in 1995 to 27.2 kg/m² in 2008, and has remained at a similar level since then (for example, it was 27.3 kg/m² in 2011). Trends in mean BMI for men and women followed a very similar pattern in this period.

The mean BMI trend for those aged 16 and over was similar to the trend for 16-64 year olds and again, the pattern was similar for both sexes. In 2003, the mean BMI for men was 27.0 kg/m², this increased slightly to 27.4 kg/m² in 2008, and has remained at a similar level to this in recent years. The equivalent figures for women ranged from 27.2 kg/m² to 27.6 kg/m², though, as with men, the highest figures have been in the more recent years.

The 2010 SHeS Report discussed the difficulties of interpreting patterns in a time series that has uneven intervals between measures.⁵ However, the latest figures appear to support the suggestion that mean BMI, and the prevalence of overweight and obesity, have begun to stabilise following the larger increases evident between the earlier years of the survey. The continued annual monitoring of these measures in the 2012-2015 period will be hugely valuable. Figure 7A, Table 7.2

7.5 ADULT BMI, BY AGE AND SEX, 2011

Table 7.3 presents the 2011 prevalence figures for the five BMI groups outlined in Section 7.2.3 (from underweight to morbidly obese) as well as the summary measures of overweight including obese (BMI of 25 kg/m² and over) and obese (BMI of 30 kg/m² and over) discussed in the previous section. In 2011, 27.7% of adults aged 16 and over were obese (27.7% of men and 27.6% of women). As Figures 7B and 7C illustrate, obesity levels varied significantly by age. There was a linear increase in prevalence from 13.4% (14.1% of men and 12.7% of women) at age 16-24 to 35.4% (35.7% of men and 35.2% women) at age 65-74, followed by a drop to 29.4% for the oldest age group (28.4% of men and 30.0% of women).

Prevalence of overweight, including obese was 64.3% among all adults in 2011 and was significantly higher in men (69.2%) than women (59.6%). The differences by age followed a similar pattern to obesity with a particularly pronounced difference between the proportion of men aged 16-24 and 25-34 that were overweight or obese (35.2% compared with 62.0%). 1.7% of men and 2.0% of women were underweight with prevalence most common among the youngest age group (8.1% of men and 7.2% of women).

The mean BMI for adults in 2011 was 27.5 kg/m² and was very similar for men (27.6 kg/m²) and women (27.5 kg/m²). Mean BMI increased significantly with age from 24.3 kg/m² for men, and 24.7 kg/m² for women aged 16-24, to a peak at age 65-74 (28.8 kg/m² for men and 28.9 kg/m² for women) before dropping slightly among the oldest age group (to 27.9 kg/m² for men and 27.5 kg/m² for women aged 75 and over). Figure 7B, Figure 7C, Table 7.3

7.6 WAIST CIRCUMFERENCE

7.6.1 Trends in waist circumference (WC) since 1995

Table 7.4 shows both the trend for mean waist circumference (WC) and for prevalence of raised WC from 1995 for adults aged 16-64, as well as figures for all adults aged 16 and over since 2003. Combined 2008/2009 and 2010/2010 data was used to allow for more detailed sub-group analysis to be carried out. Since 1995 there has been a steady increase in the mean WC of men aged 16-64 from 90.2 to 95.3 cm in 2008/2009 and 95.1 cm in 2010/2011. Over this same period there was an even greater increase in the mean WC for women, rising from 78.5 cm in 1995 to 87.2 cm in 2008/09 and 87.9 cm in 2010/2011.

The figures for all adults aged 16 and over since 2003 confirm this upward trend. Between 2003 and 2010/2011 there was a significant increase in mean WC for men and women aged 16 and over (from 95.3 cm to 96.3 cm for men and from 86.3 cm to 89.0 cm for women). However, while the overall trend has been one of increase, between 2008/2009 and 2010/2011 there was no significant change in mean WC for either men or women.

Since 1995, there has also been a steady increase in the proportion of men and women with a raised WC (greater than 102 cm for men and greater than 88 cm for women). The greatest increases occurred between 1995 and 2008/2009 with at least a doubling in the proportion of men and women aged 16-64 with a raised WC in this period (from 14.3% to 29.2% in men, and from 19.1% to 42.0% in women). The equivalent figures in 2010/2011 were 28.1% for men, and 45.5% for women.

The figures for all adults (aged 16 and over) since 2003 also show an increase in waist measurements over time, but whereas the prevalence of raised WC in men increased between 2003 and 2008/2009 and then stabilised in 2010/2011 (27.9%, 33.0% and 31.7%, respectively), for women it continued to increase (38.9%, 45.3% and 49.1%, respectively). Table 7.4

7.6.2 Waist circumference by age and sex, 2010 and 2011 combined

Mean waist circumference (WC) and prevalence of raised WC for adults aged 16 and over for 2010/2011 are shown in Table 7.4. Mean WC was 96.3 cm in men and 89.0 cm in women. There were significant differences in mean WC by age, with a linear increase up until age 55-64 for both sexes. For men, it ranged from 83.9 cm in those aged 16-24 to above 100 cm in those aged 55-64 and over (101.2 cm -103.2 cm). Among women, WC increased from 80.6 cm in the youngest age group to 93.2 cm for those aged 55-64, and then dipped slightly for the oldest group (91.9 cm).

Women were more likely than men to have a raised WC (49.1% compared with 31.7%) and, as Figure 7D illustrates, this was true across all age groups. As with mean WC, the prevalence of raised WC also increased significantly with age. 9.2% of men aged 16-24 had a raised WC and, with the exception of a blip in men aged 65-74, this increased steadily to 54.6% of those aged 75 and over. For women, prevalence increased from 26.5% of women in the youngest age group to 66.4% of those aged 65-74 before dropping to 56.0% for women aged 75 and over. Figure 7D, Table 7.4

7.7 DISEASE RISK BASED ON BMI AND WAIST CIRCUMFERENCE

7.7.1 Disease risk by age and sex, 2008-2011 combined

As described in Section 7.2.5, the WHO suggests that BMI and waist measures used in combination can provide a better estimate of adults' risk of disease. The SIGN guidelines¹⁰ on obesity management set out five risk categories: no increased risk, increased risk, high risk, very high risk and extremely high risk. Waist circumference (WC) determines the risk level (increased, high or very high) for people with a BMI between 25 and less than 35 kg/m², with a higher risk assigned to people with a higher WC. The risk level (very high and extremely high) for people with BMI levels of 35 kg/m² and above depends on BMI, regardless of WC. The inset table below and Table 7.4 show the proportions of adults in Scotland in the 2008-2011 period who were estimated to be in each of these risk categories, based on the BMI and waist measurements collected in the survey.

Risk level	Men	Women
	%	%
No increased risk	46.2	38.3
Increased risk	18.1	14.5
High risk	12.1	18.7
Very high risk	20.9	23.3
Extremely high risk	1.4	3.4

In addition to the aggregated health risk status figures for adults shown in the inset table above, a breakdown of risk status within each BMI group based on WC is also presented in Table 7.5.

The SIGN guidelines do not explicitly assign a risk status to people with a normal BMI and high or very high WC. However, in line with the advice in SIGN that this group of people can be at increased risk of some diseases, the small proportions of men (0.1%) and women (2.0%) with a normal BMI and very high WC were placed in the increased risk group.²⁴

Risk status varied by both sex and age. Men, for example, were more likely than women to fall into the no increased risk group (46.2% compared with 38.3%). 16-24 year olds were most likely to be at no increased risk of disease (72.9% and 59.7% for men and women respectively). The proportions of men in this risk group decreased with age until age 55-64 at which point it flattened out (27.0%-28.5%). For women, the decrease continued until the age of 65-74 (23.9%), before increasing again to 30.2%.

Based on their BMI and WC, 18.1% of men and 14.6% of women were classified as being at increased risk of disease. Men aged 45 and over and women aged 25 - 44 were most likely to have increased risk status while those in the youngest age group (16-24 year olds) stood out as being much less likely than other age groups to be classified as such (7.1% and 6.9% for men and women aged 16-24 respectively).

Women were more likely than men to fall into the high risk group (18.7% compared with 12.1%). For both sexes, the proportion at high risk increased steadily with age with 4.2% of men and 10.7% of women aged 16-24 were at high risk compared with 25.0% and 26.5% respectively for those aged 75 and over.

Around a fifth (20.9%) of men and a quarter (23.3%) of women were classified as being at a very high risk of disease with men aged 55-64 (32.5%) and women age 55-74 (32.2-32.3%) most likely to be classified as such. Few were classified as being at extremely high risk (1.4% of men and 3.4% of women) and while this did not vary greatly by age among men, women aged 45-54 and 65-74 were more likely to be at extremely high risk (5.1% and 5.0% respectively) than women of other ages.

The combined prevalence of those at high (or greater) risk (defined as high, very high or extremely high risk) is also shown by age and sex in Table 7.5. As the figures for the separate risk categories discussed above indicated, women were more likely than men to be at high (or greater) risk of disease (45.4% compared with 34.4%), and this was true at all ages. Based on the preceding discussions of the BMI and waist measurement results, this difference in disease risk is largely due to the prevalence of increased WC being higher in women than men. Table 7.5

According to their BMI, a significantly higher proportion of men (42.6%) than women (33.7%) were overweight (BMI 25 to <30). There were however, some striking differences in the risk status of men and women in this group. Despite having a BMI that classified them as being overweight, when examined in combination with WC, a significant proportion of overweight men (15.7%) were at no increased risk of disease. The equivalent figure for overweight women was just 4.1%. Conversely, half of overweight women were classified as being at high risk; almost double the proportion of overweight men that fell into this category. This delineation of health risk illustrates the public health importance of overweight status, particularly among women, as well as obesity.

Everyone who was obese was classified as increased risk or above. The proportion of obese men and women at increased risk was very small (just 0.4% for men and 0.1% for women). Figure 7E, Figure 7F, Table 7.5

7.7.2 Disease risk by socio-demographic factors, 2008-2011 combined

Tables 7.6 to 7.8 present results for risk status by socio-economic classification (NS-SEC of the household reference person), equivalised household income and the Scottish Index of Multiple Deprivation (descriptions of each of these measures are available in the Glossary at the end of this volume) for the combined 2008-2011 samples. In addition to presenting the figures for all of the health risk categories separately (from no increased risk to extremely high risk) the tables also present summary rows both for those classified as at high (or greater) risk, and those at very / extremely high risk.

To ensure that the comparisons presented in this section are not confounded by the different age profiles of the sub-groups, the data have been age-standardised (for a description of age-standardisation please refer to the Glossary). On the whole, the differences between observed and age-standardised percentages are small. Therefore, the percentages and means presented are the standardised ones only.

Socio-economic classification (NS-SEC)

There was a significant association between health risk category and NS-SEC, but with no clear pattern. Men in lower supervisory and technical households were the most likely to be at no increased risk (51.8%), and, along with those in professional and managerial households, were the least likely to be at a high (or greater) risk (32.6%). The pattern was a little different for women. Those in professional and managerial households were the most likely to be in the no increased risk group (43.7%), while those in intermediate, and in semi-routine and routine households, were the least likely to (32.6%). Women in semi-routine and routine households were also the most likely to be in the high (or greater) risk group (52.1%), with those in professional and managerial households the least likely (41.0%) to be. Table 7.6

Equivalised household income

Health risk category varied by equivalised household income, but again with different patterns for men and women. Men living in households in the 4^th income quintile were the most likely to be in the high (or greater) risk group (42.7%), and in the very / extremely high risk group (27.7%), and were least likely to be at no increased health risk (36.3%). However, there was no clear pattern here as those in the 3^rd income quintile had the lowest risk profile. The pattern for women was clearer: the proportion who were at no increased risk declined between the 1^st and 4^th income quintiles (from 44.1% to 34.2%), and was a little higher again for women in the 5^th (lowest) quintile (36.5%). Conversely, the proportion of women in the high (or greater) risk group increased between the 1^st and 4^th quintiles (from 40.6% to 52.7%), and then declined (to 48.8%). The pattern for the very / extremely high risk group was similar to that for the high (or greater) risk group. Table 7.7

Scottish Index of Multiple Deprivation (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 three intermediate quintiles. The second contrasts the most deprived 15% of areas with the 85% least deprived. Note that while SHeS was designed to provide robust data for the SIMD 15% areas after four years of data had been collected and combined (2008-2011), this was for the main interview sample and therefore does not apply to the nurse sub-sample which the figures in Table 7.8 and discussion below are based on.

Table 7.8 shows estimates of being in the various health risk categories by SIMD. There was some variation in risk levels across deprivation quintiles, and as with income, the pattern was slightly clearer for women than for men.

Men in the least deprived quintile were least likely to be at risk of obesity related disease - 49.1% had no increased risk, compared to 44.7%-46.6% of those in the remaining four quintiles. The patterns for the high (or greater) risk group and the very / extremely high risk group, were similar, but rather inconsistent. For example, men in the least deprived quintile and in the 3^rd quintile were equally likely to be in the high (or greater) risk group (31.0%-31.6%), while men in the most deprived quintile were the most likely to be in the high (or greater) risk group (38.8%). As Figure 7G shows, there was a more obvious gradient in the association between risk profile and area deprivation among women. The proportion at no increased risk generally declined as deprivation increased (from 45.3% in the least deprived quintile to 29.8% in the most deprived). Conversely, the proportion in the high (or greater) risk group generally increased in line with deprivation, while the proportion of women in the very / extremely high risk group doubled between the least and most deprived quintiles (from 17.7% to 35.6%).

As the quintile patterns suggest, the difference between the health risk profiles of people living in the 15% most deprived areas in Scotland and the rest of the country was more pronounced for women than for men. For example, the proportion of men in the 15% most deprived areas that were at no increased risk was similar to that for the rest of Scotland (47.5% and 46.1%, respectively). In contrast, there was a 10 percentage point difference between these groups for women (29.9% and 40.0%, respectively). Similar magnitudes of difference were seen across the other risk groups. Figure 7G, Table 7.8

7.8 FACTORS ASSOCIATED WITH HIGH (OR GREATER) DISEASE RISK

Multivariate logistic regression was used to examine the independent effect of a range of socio-demographic and behavioural factors associated with adults' disease risk. The classification, endorsed by SIGN in their guideline on obesity,¹⁰ has been use in this analysis. It uses combination of both BMI and WC to letter estimate an individuals risk of conditions like type 2 diabetes hypertension and CVD risk. A fuller discussion of the classification of disease risk used in this analysis can be found in Sections 7.2.5 and 7.7.1.

The regression explored factors independently associated with high (or greater) risk of disease. High (or greater) is defined as those classified as at high, very high or extremely high risk according to the SIGN classification.¹⁰ In the discussion that follows this group is referred to as 'high' risk.

The factors investigated included a number of the behavioural characteristics explored in other chapters in this report, such as cigarette smoking, physical activity and alcohol consumption, as well as the key socio-demographic factors of age, SIMD, equivalised household income and both parental and household NS-SEC. Regressions models were run on combined 2008-2011 data for men and women separately.

The odds ratios of being at high risk of disease are presented in Table 7.9. In these analyses, the odds of a reference group (shown in the table with a value of 1) are compared with that of the other categories for each of the individual factors. In this example, an odds ratio of greater than 1 indicates that the group in question has increased odds of having high risk of disease compared with the reference category, and an odds ratio of less than 1 mean they have decreased odds. By simultaneously controlling for a number of factors, the independent effect each factor has on the variable of interest can be established. For more information about logistic regression models and how to interpret their results see the glossary at the end of this volume.

The factors found to be associated with high disease risk for both men and women were: age, economic status and physical activity. Additionally, educational attainment, marital/partnership status and self-assessed health were significant factors for men while SIMD, parental NS-SEC, smoking status and longstanding illness were also significant for women.

When compared with women aged 16-24, women aged 45 and over had increased odds of being at high risk of disease (odds ratios of 1.78 to 3.06). The odds of being at high risk of disease were highest for those aged 55-64 (3.06 times higher than for the youngest age group). Overall, age was associated significantly with high disease risk for men, but the nature of the relationship was not clear.

For both men and women, economic status was independently associated with being at high risk of disease but the nature of the relationship differed slightly. Men in education had lower odds of being at high risk than those in the reference group - men in paid employment, self-employed, on government training or doing something else (odds ratio of 0.10). For women, those who were retired or looking after home/family had decreased odds when compared to the reference group (odds ratio of 0.66).

Physical activity levels were also associated with disease risk for both men and women. Three levels of physical activity were examined: high (meeting the recommended level of 30 minutes or more at least 5 days a week); medium (30 minutes or more on 1 to 4 days a week); and low (fewer than 30 minutes of activity a week). Compared with those in the high physical activity group, those with medium and low activity levels had significantly increased odds of being classified as at high risk of disease with those who were least active (low) having the greatest odds (the odds ratios for men were 1.89 for the medium activity level group and 2.41 for the low activity group, equivalent figures for women were 1.72 and 2.56 respectively).

Educational attainment was associated with being at high risk of disease for men: with those with no qualifications or who did not supply information of their education having significantly higher odds of being at increased health risk than those with degree or higher qualifications (odds ratio of 1.64). Marital status was also a significant factor for men, with single, separated/divorced and widowed men all having lower odds of being at high risk when compared with men that were married or living as married (odds ratios of 0.63, 0.52 and 0.54 respectively).

When compared with men who had never smoked cigarettes, those who smoked had decreased odds of being at high disease risk (odds ratio of 0.66). Overall, self-assessed health was also significantly associated with high disease risk among men but with no clear pattern (p=0.027). Neither smoking status nor self-assessed health were significant factors for women.

For women, SIMD was also associated with being at high risk of disease. Those living in the 2^nd, 4^th and 5^th (most deprived) quintiles had significantly increased odds of being at high risk when compared with those living in the least deprived quintile (odds ratios of 1.43, 1.74 and 1.93 respectively).

Parental socio-economic classification (NS-SEC) was also independently associated with high risk of disease for women. Women with semi-routine or routine backgrounds had significantly increased odds of being at high disease risk when compared with those whose parents worked in managerial and professional occupations (odds ratio of 1.34).

Overall, cigarette smoking status and presence of a longstanding illness were significantly associated with being at high risk of disease for women but the nature of these relationships was unclear (p=0.016 and p=0.036, respectively). Table 7.9