Accessibility statement

Re-estimating the social gradient in the health effects of marginal health care expenditure

Thursday 10 February 2022, 11.15AM to 12.15pm

Speaker(s): Katja Grasic, CHE

Background: Several studies have produced direct regression-based estimates of the overall health effects of marginal healthcare expenditure in England (e.g. Andrews et al 2017). By combining this with data on social gradients in healthcare utilisation, Love-Koh et al (2020) produced indirect estimates of the social gradient in the health effects of marginal healthcare expenditure. 

Objective: Our objectives are three-fold. First, directly to estimate the social gradient in the mortality effects of marginal healthcare expenditure, using hospital allocations data. Second, to calculate effects in terms of quality adjusted life expectancy (QALE). Third, to re-estimate the social gradient in the health effects of marginal healthcare expenditure and compare this with previous estimates.  

Data and Methods: Our analysis employs planning area level (“Clinical Commissioning Group”) population and mortality data for 2018, Index of Multiple Deprivation scores for 2019, and per-head financial allocations for hospital general and acute services together with components of the corresponding allocation formula including market forces factor, distance to target and prescription age index for 2018. We use an instrumental variable approach with mortality as the dependent variable. We use components of the formula to instrument hospital expenditure and stratify within-planning-area mortality by five neighbourhood deprivation quintile groups. In next steps we will assume morbidity effects are proportional to mortality effects and use the Sullivan method to calculate effects on quality adjusted life expectancy using 2018 Health Survey for England EQ-5D-5L data. 

Preliminary Results: In our preliminary analysis, we estimate that an increase in hospital expenditure of 1% decreases the age-sex adjusted mortality rate by -523 in 100,000 (95% CI: -1,148 to 101) for the most deprived group and -388 (95% CI: -691 to -86) for the least deprived group. This suggests the social gradient in health effects is shallower than previously thought, and may even be non-existent. This would imply that simply providing health services with additional funding is not enough to reduce health inequality – targeting funding towards specific activities expected to reduce health inequality may be necessary.

Join Zoom Meeting Meeting ID: 962 7470 4337 Passcode: 534466

Location: Zoom Presentation (not recorded)

Who to contact

For more information on these seminars, contact:
Alfredo Palacios
alfredo.palacios@york.ac.uk
Shainur Premji
shainur.premji@york.ac.uk

If you are not a member of University of York staff and are interested in attending a seminar, please contact
alfredo.palacios@york.ac.uk 
or
shainur.premji@york.ac.uk 
so that we can ensure we have sufficient space

Economic evaluation seminar dates

  • Tuesday 28 November 2023
  • Thursday 14 December 2023