Health Inequalities Dataset

Health Inequalities Dataset

September 2015

This paper begins to set out the scale of health inequality in Grampian, Aberdeen City, Aberdeenshire and Moray. Further indicators will be added over time.

The Health Inequalities Dataset Group

In July 2014 a group was established to consider the best tools for measuring and monitoring the gap in health and wellbeing outcomes and lifestyle choices on an annual basis, with respect to deprivation in Grampian, and the three CHPs within Grampian, to help inform priorities for reducing the outcomes gap.

The group comprises: Dr Linda Leighton-Beck (Head of Social Inclusion, NHS Grampian); Kim Penman (Health and Wellbeing Lead, Aberdeenshire CHP); Chris Littlejohn (Head of Health Improvement, NHS Grampian); Linda Smith (Public Health Lead, Aberdeen City CHP); Tracey Gervaise (Health and Wellbeing Lead, Moray CHP), and members of Health Intelligence, NHS Grampian.

A number of documents were considered at the initial meeting, including a report from Dr Forster: The Intelligent Board 2009 Commissioning to reduce inequalities[i], and a ScotPHO guidance document[ii] on methods for measuring inequalities.

The first document helped to influence the initial choice of indicators. The second document informed the decision to use the relative index of inequality (RII) where possible, as this has a number of advantages:

·  It is a relative measure

·  It has a straightforward interpretation using non-technical language

·  A single value can be quoted

·  It can be calculated using formulae in Excel

Explanation of the Relative Index of Inequality (RII)

To explore the relationship between health and deprivation, the deprivation gradient for a set of indicators has been calculated using the RII. The RII uses the gradient of the best-fit linear regression against deprivation, and adjusts for population size in each deprivation quintile. It is then divided by the average rate for the area concerned. In so doing, the index becomes relative (to the average) and is not sensitive to the magnitude of the prevalence, thereby allowing the less prevalent health conditions to be compared directly to the more prevalent conditions.

In the following interpretation, SIMD1 is the most deprived quintile, and SIMD5 is the least deprived quintile. For A&E attendance rates and breast feeding rates, health board deprivation quintiles have been used. For all other indicators, local CHP deprivation quintiles have been used.

Contents

A&E Attendance Rates

Breast Feeding at 6-8 Weeks

Dental Caries in Primary 1 Children

Dental Caries in Primary 7 Children

Teenage Pregnancy

Smoking in Pregnancy

Adult Smoking Status

The following indicators are described in separate papers:

Hospital Preventable Admissions [iii]

Amenable Deaths[iv]

A&E Attendance Rates (2014)

Aberdeen City

There is a clear relationship between A&E attendance rates and deprivation in Aberdeen City. The highest rate is in SIMD1 at 295 per 1000 population and the lowest in SIMD4 and SIMD5 at 153.1 per 1000 population.

According to the regression equation the range between the most and least deprived is 1.02 times the average of 214.6 per 1000 population.

The differences observed in attendance rates across the quintiles are not due to differing age profiles as it can be seen that crude rates are practically identical to age standardised rates in each quintile.

The same pattern is evident in each of the past five years with SIMD1 having a substantially higher attendance rate and SIMD5 having a higher attendance rate than SIMD4.

Aberdeenshire

No relationship between attendance rates and deprivation can be seen in Aberdeenshire. Rates are highest in SIMD3 and lowest in SIMD1. This is because the majority of the most deprived areas are in Peterhead and Fraserburgh where attendances are largely at Minor Injuries Units. Due to proximity to Aberdeen Royal Infirmary, Westhill and Portlethen, areas with little deprivation, have the highest attendance rates at A&E.

According to the regression equation the range between the most and least deprived is only 0.11 times the average of 109.1 per 1000 population

The differences observed in attendance rates across the quintiles are not due to differing age profiles as it can be seen that crude rates are practically identical to age standardised rates in each quintile.

The pattern of attendances has been the same in each year since 2010 with SIMD1 having the lowest attendance rate and SIMD3 and SIMD 5 the highest rates.

Moray

Attendance rates in Moray are much higher in SIMD1 and SIMD2 but there is little difference in attendance rates amongst the three least deprived quintiles.

According to the regression equation the range between the most and least deprived is 0.55 times the average of 235.8 per 1000 population.

The differences observed in attendance rates across the quintiles are not due to differing age profiles as it can be seen that crude rates are practically identical to age standardised rates in each quintile.

The same pattern in attendance rates can be seen in most years since 2010 with SIMD1 having the highest attendance rate and SIMD3 the lowest.

Breastfeeding at 6-8 Weeks (2014/15)

Aberdeen City

There is a strong relationship between deprivation and breastfeeding rates with SIMD1 having the lowest exclusive breastfeeding rate at 6-8 weeks, though SIMD3 has the highest rate.

According to the regression equation the range between the most and least deprived is 0.71 times the average of 37.4%.

Aberdeenshire

There is a clear relationship between deprivation and breastfeeding rates in Aberdeenshire with the lowest rates in SIMD1 and increasing in each quintile up to SIMD4. SIMD5 however records a much lower rate than SIMD4.

According to the regression equation the range between the most and least deprived is 0.58 times the average of 30.9%.

Moray

There is also a clear relationship between deprivation and breastfeeding rates in Moray. As in Aberdeenshire it can be seen that SIMD5 has a lower rate of exclusive breastfeeding than SIMD4.

According to the regression equation the range between the most and least deprived is 0.45 times the average of 30.9%.

Dental Caries Primary 1 (2012/13)

Aberdeen City

The regression line gives an exceptionally good fit illustrating a clear relationship between deprivation and dental caries. 50.8% of Primary 1 children in SIMD1 have dental caries compared to only 12.7% in SIMD5.

According to the regression equation the range between the most and least deprived is 1.44 times the average of 32.4%.

Aberdeenshire

For Aberdeenshire the regression line is not such a good fit as with Aberdeen City. However, the percentage with dental caries in SIMD 1 and 2 is still much higher but there is little difference between the other three quintiles.

According to the regression equation the range between the most and least deprived is 1.08 times the average of 27.7%.

Moray

In Moray the percentage with Dental caries in SIMD1 is much higher but there is no difference in the percentages between SIMD2 and SIMD4.

According to the regression equation the range between the most and least deprived is 0.34 times the average of 37.4%.

Dental Caries Primary 7 (2012/13)

Aberdeen City

Again there is a clear relationship between dental caries and deprivation though the drop is not as marked as with Primary 1 pupils.

According to the regression equation the range between the most and least deprived is 0.5 times the average of 55.3%.

Aberdeenshire

There is a clear relationship between dental caries and deprivation with 76.1% in SIMD1 having dental caries compared to 53.8% in SIMD5.

According to the regression equation the range between the most and least deprived is 0.44 times the average of 61.7%.

Moray

No relationship can be seen between dental caries and deprivation in Moray. SIMD2 has the lowest rate of dental caries and SIMD3 the highest.

The regression equation confirms that the rate among those at the bottom is no different from those at the top (the range between the most and least deprived is only 0.01 times the average of 55.5%).

Teenage Pregnancy (2010-14)

Aberdeen City

There is a clear relationship between deprivation and teenage pregnancy rate with an average of 91.7 per 1000 population per annum in SIMD1 compared to 18.7 per 1000 population in SIMD4. The least deprived quintile however has a slightly higher rate.

According to the regression equation the range between the most and least deprived is 1.96 times the average of 43.4%.

Aberdeenshire

Teenage pregnancy rates follow the same pattern as Aberdeen City though rates are lower in each quintile. Again the rate does not drop between SIMD4 and SIMD5.

According to the regression equation the range between the most and least deprived is 1.27 times the average of 30.0%.

Moray

Teenage pregnancy rates in Moray again follow the same pattern but with rates between those of Aberdeen City and Aberdeenshire. Here the rate drops between SIMD4 and SIMD5 but the difference is slight compared to the other quintiles.

According to the regression equation the range between the most and least deprived is 1.58 times the average of 37.7%.

Smoking in Pregnancy (2014/15)

Aberdeen City

Smoking rates amongst pregnant women drop in successive quintiles from 27.1% in the most deprived quintile to 3.1% in the least deprived.

According to the regression equation the range between the most and least deprived is 2.09 times the average of 14.9%.

Aberdeenshire

The pattern in Aberdeenshire is the same as in Aberdeen city, smoking rates dropping from 26.0% in the most deprived quintile to 7.1% in the least deprived.

According to the regression equation the range between the most and least deprived is 1.68 times the average of 14.1%.

Moray

In Moray smoking rates dropped from 31.8% in the most deprived quintile to 10.4% in the least deprived. It is noticeable that the smoking rate in SIMD2 is lower than in both SIMD3 and SIMD5.

According to the regression equation the range between the most and least deprived is 1.09 times the average of 20.6%.

Adult Smoking Status

Data available from the Scottish Health Survey[v] allow analysis at CHP level for two combined years for Aberdeen City and Aberdeenshire and four combined years for Moray. Income quintile has been used to rank the respondents.

Aberdeen City

In Aberdeen City for combined years 2012 and 2013, current male smoking rates dropped from 52.6% in the lowest income quintile to 13.7% in the highest income quintile. According to the regression equation the range between the lowest and highest income is 1.72 times the average of 24.8%.

Meanwhile the female smoking rates dropped from 30.0% in the lowest income quintile to 10.5% in the highest income quintile. There is a stronger correlation between smoking and income for female smokers than male smokers. According to the regression equation the range between the lowest and highest income is 1.36 times the average of 20.6%, so the actual and relative gap is less for females than males.

Aberdeenshire

In Aberdeenshire for combined years 2012 and 2013, current male smoking rates dropped from 25.7% in the lowest income quintile to 12.8% in the highest income quintile, with strong correlation between smoking and income. According to the regression equation the range between the lowest and highest income is 1.04 times the average of 24.8%.

Meanwhile the female smoking rates dropped from 31.4% in the lowest income quintile to 11.8% in the highest income quintile. There is a weaker correlation between smoking and income for female smokers than male smokers. According to the regression equation the range between the lowest and highest income is 1.02 times the average of 16.5%, so a similar relative gap to male smokers.

Moray

In Moray for combined years 2010 to 2013, current male smoking rates varied from 24.0% in the second highest income quintile to 16.8% in the highest income quintile, with no correlation between smoking and income. The average rate was 20.4%.

Meanwhile the female smoking rates dropped from 23.1% in the lowest income quintile to 10.7% in the highest income quintile. According to the regression equation the range between the lowest and highest income is 0.66 times the average of 18.9%.

Richard Slessor

Health Intelligence

NHS Grampian

4 September 2015

Page 2 of 19

[i]i Dr Forster (2009) The Intelligent Board 2009 Commissioning to reduce inequalities available from http://drfosterintelligence.co.uk/thought-leadership/intelligent-board

[ii]

ii ScotPHO (2007) Measuring Socio-Economic Inequalities in Health: A Practical Guide available from http://www.scotpho.org.uk/downloads/scotphoreports/scotpho071009_measuringinequalities_rep.pdf

[iii] NHS Grampian Health Intelligence (2015) Hospital Preventable Admissions

[iv] NHS Grampian Health Intelligence (2015) Amenable Deaths

[v] UK Data Service (2015) Scottish Health Survey data on smoking status available from www.ukdataservice.ac.uk (with appropriate permission).