Using IMD in Statistical Models of Mortality It is common in the world of health to incorporate measures of socioeconomic deprivation into statistical models. One commonly used proxy-measure of deprivation is the UK's Indices of Multiple Deprivation (IMD) score.
As the name suggests, this score is derived from several different indices:


*

*Income Score

*Employment Score

*Health Deprivation And Disability Score

*Education Skills And Training Score

*Barriers To Housing And Services Score

*Crime And Disorder Score

*Living Environment Score


What has been playing on my mind recently is the Health Deprivation and Disability domain (described below). The IMD score includes measures of mortality and morbidity, but is commonly used as an explanatory variable in statistical models which look at health measures such as ... mortality and morbidity.
An example of this can be seen here: Socioeconomic indicators of health inequalities and female mortality: a nested cohort study within the United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS)
Is this a flawed method? Should the Health Deprivation And Disability Score domain be excluded from analyses of variables which may overlap with this domain?


Health Deprivation and Disability Domain
This domain measures premature death and the impairment of quality of
  life by poor health. It considers both physical and mental health. The
  domain measures morbidity, disability and premature mortality but not
  aspects of behaviour or environment that may be predictive of future
  health deprivation. Four indicators are used to calculate this domain:
  
  
*
  
*Years of Potential Life Lost – an age and sex standardised measure of    premature death
  
*Comparative Illness and Disability Ratio – an age and sex standardised measure of morbidity and disability
  
*Measures of acute morbidity – an age and sex standardised rate of emergency admissions to hospital
  
*Proportion of adults under 60 suffering from mood or anxiety disorders – a modelled indicator for the proportion of adults suffering from mood and anxiety disorders.
  
  
  The English Indices of Deprivation 2010

 A: Having asked around some colleagues, I have been pointed towards this paper by Adams & White (2006): Removing the health domain from the Index of Multiple Deprivation 2004—effect on measured inequalities in census measure of health
Whilst this study has some limitations in terms of the variables investigated (IMD 2004, long-term limiting illness and less-than-good self-reported health), it does at least present some degree of investigation into the topic.
The paper makes the following recommendation:

Removing the health domain from IMD 2004 did require some technical
  expertise and time. Although our findings suggest that removing the
  health domain from IMD 2004 is unnecessary for routine public health
  analyses using census measures of health, mathematical coupling and
  the associated problems in interpreting results remain a potential
  issue when using IMD 2004 to investigate patterns in other measures of
  health. In the context of socioeconomic inequalities in health,
  removing the health domain from IMD 2004 probably represents best
  practice.

