 |
 |
 |
 |
||
 |
||
| Home | Command Papers | House Papers | Departmental Papers | Search | Site Map | Contact Us | Links | ||
 |
|
|
|
|
|
HSE - Methodology and Documentation
|
| BMI (kg/m2) | Description |
| 18.5 or less | Underweight |
| Over 18.5 to 25 | Desirable |
Over 25 to 30 |
Overweight |
| Over 30 | Obese |
In those with a BMI over 40, the condition is defined as 'morbid obesity'. Although the BMI calculation method is the same, there are no fixed BMI cut-off points defining overweight and obesity in children. Instead, overweight and obesity are defined using other methods, including age and sex specific BMI cut-off points, or BMI percentiles cut-offs based on reference populations.
Cardiovascular disease
Informants were classified as having cardiovascular disease (CVD) if they reported ever having any of the following conditions diagnosed by a doctor: angina, heart attack, stroke, heart murmur, irregular heart rhythm, 'other heart trouble'. In contrast with the definition used in previous HSE reports, high blood pressure and diabetes are not included in this definition, since they are risk factors for CVD and are dealt with in separate chapters. For the purpose of this report, informants were classified as having a particular condition only if they reported that the diagnosis was confirmed by a doctor. No attempt was made to assess these self-reported diagnoses objectively. There is therefore the possibility that some misclassification may have occurred, because some informants may not have remembered (or not remembered correctly) the diagnosis made by their doctor.
Cholesterol (Total and HDL)
Cholesterol is a fat-like substance (lipid) that is present in cell
membranes and is a precursor of bile acids and steroid hormones. Cholesterol is essential for the body in small amounts. It is made in the liver and some is obtained from the diet. Serum total cholesterol concentration is positively associated with the risk of coronary heart disease (CHD).
In this report, raised total cholesterol has been defined as >=5.0 mmol/l. At the time of the last HSE report on CVD (HSE 1998), the level was >=6.5 mmol/l. This has been taken account of in comparisons over time.
In a normal individual, high density lipoprotein (HDL) constitutes approximately 20-30% of total plasma cholesterol. Studies have demonstrated a strong direct relationship between coronary heart disease and low HDL-cholesterol. HDL-cholesterol was considered low at a level of less than 1.0 mmol/l.
See also LDL-cholesterol.
Cotinine
Cotinine is a metabolite of nicotine. It is one of several biological markers that are indicators of smoking. In this survey, it was measured in saliva (from all aged 4-15 and from a subsample of those aged 16 and over). It has a half-life in the body of between 16 and 20 hours, which means that it will detect regular smoking (or other tobacco use such as chewing) but may not detect occasional use if the last occasion was several days ago. Anyone with a salivary cotinine level of 15 nanograms per millilitre or more is highly likely to be a tobacco user.
C-reactive protein
C-reactive protein (CRP) is the major protein indicating inflammation activity in acute illness in humans. It is also a marker of cardiovascular risk. No recommendations for CRP thresholds appear in the literature so quintile distributions have been presented in this report. The categories of CRP defined on the basis of the quintile distribution, separately for men and women, are:
| Men | Women | |
Bottom quintile |
≤ 0.5 |
≤ 0.5 |
2nd quintile |
0.6-1.0 |
0.6-1.2 |
3rd quintile |
1.1-1.9 |
1.3-2.4 |
4th quintile |
2.0-3.7 | 2.5-4.9 |
| Top quintile | > 3.7 |
> 4.9 |
Creatinine This is excreted in urine and unlike sodium and potassium is relatively stable over time. This was analysed using a spot urine sample in HSE 2003. Therefore in the analysis of dietary sodium and potassium the ratio of sodium to creatinine and the ratio of potassium to creatinine is analysed. See also Urine, Sodium, Potassium
Equivalised household income
Income was not included in the Health Survey series until 1997. Making precise estimates of household income, as is done for example in the Family Resources Survey, requires far more interview time than was available in the Health Survey. Household income was thus established by means of a card (see Appendix A) on which banded incomes were presented. Information was obtained from the household reference person (HRP) or their partner. Initially they were asked to state their own (HRP and partner) aggregate gross income, and were then asked to estimate the total household income including that of any other persons in the household. Household income can be used as an analysis variable, but there has been increasing interest recently in using measures of equivalised income that adjust income to take account of the number of persons in the household. Methods of doing this vary in detail: the starting point is usually an exact estimate of net income, rather than the banded estimate of gross income obtained in the Health Survey. The method used in the present report was as follows. It utilises the widely used McClements scoring system, described below.
1. A score was allocated to each household member, and these were added together to produce an overall household McClements score. Household members were given scores as follows.
| First adult (HRP) | 0.61 |
| Spouse/partner of HRP | 0.39 |
Other second adult |
0.46 |
Third adult |
0.42 |
Subsequent adults |
0.36 |
Dependant aged 0-1 |
0.09 |
Dependant aged 2-4 |
0.18 |
Dependant aged 5-7 |
0.21 |
Dependant aged 8-10 |
0.23 |
Dependant aged 11-12 |
0.25 |
Dependant aged 13-15 |
0.27 |
Dependant aged 16+ |
0.36 |
2. The equivalised income was derived as the annual household income divided by the McClements score.
3. This equivalised annual household income was attributed to all members of the household, including children.
4. Households were ranked by equivalised income, and quintiles q1- q5 were identified. Because income was obtained in banded form, there were clumps of households with the same income spanning the quintiles. It was decided not to split clumps but to define the quintiles as 'households with equivalised income up to q1', 'over q1 up to q2' etc.
5. All individuals in each household were allocated to the equivalised household income tertile to which their household had been allocated. Insofar as the mean number of persons per household may vary between tertiles, the numbers in the quintiles will be unequal. Inequalities in numbers are also introduced by the clumping referred to above, and by the fact that in any sub-group analysed the proportionate distribution across quintiles will differ from that of the total sample.
Reference: McClements D. Equivalence scales for children. Journal of Public Economics 1977; 8: 191-210.
Fibrinogen
Fibrinogen is a soluble protein involved in the blood clotting mechanism. Prospective population studies have established that fibrinogen is an independent predictor for ischaemic heart disease and stroke.
Reference: Maresca G, Di Blasio A, Marchioli R, Di Minno G. Measuring plasma fibrinogen to predict stroke and myocardial infarction. Arterioscler Thromb Vasc Biol 1999; 19:1368-1377.
Geometric mean
The geometric mean is a measure of central tendency. It is sometimes preferable to the arithmetic mean, since it takes account of positive skewness in a distribution. The geometric mean of a continuous variable is calculated by taking the antilog of the mean of the logged original values. See also Arithmetic mean.
GHQ12
The General Health Questionnaire (GHQ12) is a scale designed to detect possible psychiatric morbidity in the general population. It was administered to informants aged 13 and above. The questionnaire contains 12 questions about the informant's general level of happiness, depression, anxiety and sleep disturbance over the past four weeks.
Reference: Goldberg D, Williams PA. User's Guide to the General Health Questionnaire. NFER-NELSON, 1988.
Glucose
Glucose, also called blood sugar, provides a source of energy for tissue cells and is the only source of energy for some cells, for example red blood cells. The liver regulates blood sugar levels by removing glucose after a meal and by glucose production during overnight fast. Insulin is the hormone that helps the glucose enter the cells. Without insulin, or with very little of it (as in Type 2 diabetes), high blood sugar levels (hyperglycemia) build up which can cause health problems in the long term such as eye disease, kidney failure and stroke. In 1999 the UK adopted the WHO guidelines for diagnosing diabetes, which recommend a fasting plasma glucose threshold of 7.0 mmol/l. Glucose levels were measured on a fasting blood sample.
Glycated haemoglobin
The percentage of glycated haemoglobin is the percentage of haemoglobin in the circulation to which glucose is bound. Glycated haemoglobin (HbA1c) concentration is an indicator of average blood glucose concentration over three months and has been suggested as a diagnostic or screening tool for diabetes. Diabetic patients with elevated glycated haemoglobin are at increased risk of microvascular and macrovascular events. Raised glycated haemoglobin has been taken as equal to or greater than 7%.
Government Office Region
Government Office Region (GOR) is the key classification system used for regional statistics. There are nine Government Office Regions in England: North East, North West, Yorkshire and the Humber, East Midlands, West Midlands, East England, London, South East and South West. The nine category system has been used since 1998, however, GOR boundaries may change from year to year as they reflect administrative boundaries.
High blood pressure
See Blood pressure.
Household
A household was defined as one person or a group of people who have the accommodation as their only or main residence and who either share at least one meal a day or share the living accommodation.
Household reference person The household reference person (HRP) is defined as the householder (a person in whose name the property is owned or rented) with the highest income. If there is more than one householder and they have equal income, then the household reference person is the eldest.
Income
See Equivalised household income.
Ischaemic heart disease
Informants were classified as having ischaemic heart disease (IHD) if they reported ever having angina or a heart attack diagnosed by a doctor.
Linear regression
Linear regression was used to investigate the independent effects of two or more factors ('independent' or 'predictor' variables) on a continuous variable ('dependent' or 'outcome' variable), such as blood pressure. The independent variables can be continuous or categorical (grouped) variables. The parameter estimates for a particular variable from a linear regression model give an estimate of the effect of that variable on the outcome variable, adjusted for all other variables in the model. For example, this was used in the analysis of blood pressure.
For a continuous independent variable, the regression coefficient is the change that is predicted in the mean of the outcome variable for a one unit change in the independent variable, adjusted for all other variables in the model.
Parameter estimates for categorical independent variables have been presented using the standard method which defines one category of a categorical independent variable as a baseline or reference category and compares all other categories to this reference category. Therefore there is no parameter estimate for the reference category and estimates for all other categories give the predicted mean difference in the outcome variable between each category and the reference category, adjusted for all other variables in the model.
95% confidence intervals were calculated for these parameter estimates. These can be interpreted as meaning there is a 95% chance that the given interval for the sample will contain the true population parameter of interest. In linear regression a 95% confidence interval which does not include zero indicates that the given parameter estimate is statistically significant.
Reference: Weisberg, S. Applied linear regression. John Wiley & Sons,
New York 1985.
LDL-cholesterol
Low density lipoprotein (LDL) cholesterol makes up 60-70% of total serum cholesterol. LDL-cholesterol is a well recognized CHD risk factor: the higher the levels of LDL-cholesterol in the blood, the greater the risk of heart disease. An LDL-cholesterol level of 3.0 mmol/l or above is referred to as 'high' LDL-cholesterol. LDL-cholesterol was measured on a fasting blood sample.
Logistic regression
Logistic regression was used to investigate the effect of two or more independent or predictor variables on a two-category (binary) outcome variable. The independent variables can be continuous or categorical (grouped) variables. The parameter estimates from a logistic regression model for each independent variable give an estimate of the effect of that variable on the outcome variable, adjusted for all other independent variables in the model.
Logistic regression models the log 'odds' of a binary outcome variable. The 'odds' of an outcome is the ratio of the probability of its occurring to the probability of its not occurring. The parameter estimates obtained from a logistic regression model have been presented as odds ratios for ease of interpretation.
For continuous independent variables, the odds ratio gives the change in the odds of the outcome occurring for a one unit change in the value of the predictor variable.
Parameter estimates for categorical independent variables have been presented using one category of the categorical variable, selected as a baseline or reference category, with all other categories compared to it. Therefore there is no parameter estimate for the reference category and odds ratios for all other categories are the ratio of the odds of the outcome occurring between each category and the reference category, adjusted for all other variables in the model.
The statistical significance of independent variables in models was assessed by the likelihood ratio test and its associated p value. 95% confidence intervals were also calculated for the odds ratios. These can be interpreted as meaning that there is a 95% chance that the given interval for the sample will contain the true population parameter of interest. In logistic regression a 95% confidence interval which does not include 1.0 indicates the given parameter estimate is statistically significant.
Longstanding illness and limiting longstanding illness
Longstanding illness was defined as an illness, disability or infirmity that had troubled the informant over a period of time or was likely to affect them over a period of time. Longstanding illnesses were coded into categories defined in the International Classification of Diseases (ICD), but it should be noted that the ICD is used mostly to classify conditions according to the cause, whereas HSE classifies according to the reported symptoms. A longstanding illness was defined as limiting if the respondent reported that it limited their activities in any way.
Mean
Means in this report are mostly Arithmetic means (the sum of the values for cases divided by the number of cases). In a few cases the Geometric mean has been used.
Median
The value of a distribution which divides it into two equal parts such that half the cases have values below the median and half the cases have values above the median.
Morbid obesity
See Body mass index.
Moving averages
The large sample sizes in the Health Survey provide the opportunity to analyse by individual years of age instead of the more usual discrete age groups. However, even with several hundred per group there is necessarily some random fluctuation that may obscure the underlying trends or patterns. The amount of such fluctuation varies from one variable to another, and depends to a considerable extent, though not solely, on the size of the sub-group to whom the question is addressed. To minimise random variation and to bring out the underlying pattern more clearly, the method of moving averages has been adopted in the construction of some graphs.
In constructing a three year moving average, the percentages shown for each of the first three age years are summed and divided by three to give the first value which is plotted against the central year. The first figure being the average of ages 16, 17 and 18 plotted against 17, the second figure the average of 17, 18 and 19 plotted against 18 and so on.
The advantage of moving averages is the removal of 'noise' in the data, and they work well when change is relatively slow and uniform. Their disadvantage is that if the real underlying pattern involves sharp changes of direction (for example, if values of a variable initially decline sharply with age, then increase to a marked peak and then decrease) moving averages will tend to iron out the turning points.
NS-SEC
The National Statistics Socio-economic Classification (NS-SEC) is a social classification system that attempts to classify groups on the basis of employment relations, based on characteristics such as career prospects, autonomy, mode of payment and period of notice. There are fourteen operational categories representing different groups of occupations (for example higher and lower managerial, higher and lower professional) and a further three 'residual' categories for full-time students, occupations that cannot be classified due to lack of information or other reasons. The operational categories may be collapsed to form a nine, eight, five or three category system. The Health Survey for England uses the five category system in which informants are classified as managerial and professional, intermediate, small employers and own account workers, lower supervisory and technical, and semi-routine and routine occupations. In analyses presented in this report it is the NS-SEC of the household reference person which is used.
Obesity
See Body mass index.
Odds ratio
See Logistic regression.
Overweight
See Body mass index.
Percentile
The value of a distribution which partitions the cases into groups of a specified size. For example, the 20th percentile is the value of the distribution where 20 percent of the cases have values below the 20th percentile and 80 percent have values above it. The 50th percentile is the median.
Potassium
The intake of potassium (Na) can be estimated by measuring urinary excretion. This was analysed in HSE 2003 using a spot urine sample. See also Urine, Sodium, Creatinine. There is an association between potassium intake and blood pressure.
p value
A p value is the probability of the observed result occurring due to chance alone. A p value of less than 5% is conventionally taken to indicate a statistically significant result (p<0.05). It should be noted that the p value is dependent on the sample size, so that with large samples differences or associations which are very small may still be statistically significant. Results should therefore be assessed on the magnitude of the differences or associations as well as on the p value itself. The p values given in this report have taken into account the weighting, clustering and stratification of the survey design.
Quintile
Quintiles are percentiles which divide a distribution into fifths, i.e., the 20th, 40th, 60th and 80th percentiles.
Region
See Government Office Region.
Social class of head of household
A social class was assigned on the basis of the occupation of the head
of household using the Registrar General's Standard Occupational Classification. Occupations are assigned to six social class categories:
| Social Class | Occupations |
| I | Professional occupations |
|
II |
Managerial and technical occupations |
|
III |
Skilled occupations |
|
(IIINM) |
(Non-manual) |
|
(IIIM) |
(Manual) |
|
IV |
Partly skilled occupations |
|
V |
Unskilled occupations |
These six social classes have been combined into two: non-manual (I, II, IIINM) and manual (IIIM, IV, V). This was used in the design and weighting of the 2003 sample but is not used in the reporting.
Social support
The perceived social support scale, originally used in the Health and Lifestyle Survey, was based on seven questions about physical and emotional aspects of social support. Informants were asked about the amount of support and encouragement they received from family friends. These questions were combined into a single scale categorising informants as having 'a severe lack', 'some lack' or 'no lack' of social support.
Reference: Cox BD et al. The Health and Lifestyles Survey. The Health Promotion Research Trust, London, 1987.
Sodium
The intake of sodium (Na) can be estimated by measuring urinary excretion. This was analysed in HSE 2003 using a spot urine sample. There is an association between sodium intake and blood pressure. See also Urine, Potassium, Creatinine.
Standardisation
In this report, standardisation refers to standardisation (or 'adjustment') by age (see Age standardisation).
Triglycerides
Triglycerides are formed in the intestine from dietary fat and appear in the blood after a fat-containing meal. There is increasing evidence that raised serum triglycerides levels are an independent risk factor for the incidence of CVD. Triglycerides levels were measured on a fasting blood sample.
Unit of alcohol
A unit of alcohol is 8 gms of ethanol, and is the amount contained in half a pint of ordinary beer or lager, or in a small glass of wine, or in a measure of spirits.
Urine analysis
A spot urine sample was collected from a subsample of adults (aged 16 and over). (For details see section 9.2.3.) This was used for the analysis of dietary Sodium, Potassium and Creatinine. Epidemiological, clinical and animal-experimental evidence shows a direct relationship between dietary sodium and potassium consumption and blood pressure (BP).
Waist-hip ratio
Waist-hip ratio (WHR) was defined as the waist circumference divided by the hip circumference, i.e. waist girth (m)/hip girth (m). WHR is a measure of deposition abdominal fat, i.e. central obesity. Unlike BMI there is no consensus on a cut-off point for WHR to define central obesity, but several have been proposed. For consistency the cut-off values from the 1998 report have been used. A raised WHR has been taken to be 0.95 or more in men and 0.85 or more in women.
Reference: Molarius A, Seidell JC. Selection of anthropometric indicators for classification of abdominal fatness - a critical review. Int J Obes 1998; 22:719-727.