This is from the ONS statistical bulletin entitled "Life Expectancy at Birth and at Age 65 by Local Areas in the United Kingdom, 2006-08 to 2010-12". It talks about possible reasons for health inequality - including this one:
One factor that has received less attention is the selective migration of healthy individuals from poorer health areas into better health areas or vice-versa. This type of migration has been shown to play a significant role in increasing or decreasing location-specific illness and mortality rates, which then consequently impact on life expectancy figures. Norman, Boyle and Rees (2005) demonstrated that the largest absolute flow within England and Wales between 1971 and 1991 was of relatively healthy people moving from more deprived into less deprived areas. The impact of this migration was to raise ill-health and mortality rates where these people originated from and lower them in the destination areas. The authors also noted that the benefit to less deprived areas was reinforced by a significant group of people in poor health who moved from less to more deprived locations.
Migration explains a lot of the variation it seems. So area-based approaches to reducing 'health inequality' may be addressing entirely the wrong target problem.
H/T Tim W