Substance misuse related hospital admissions, costs and treatment outcomes: econometric analysis of administrative data for England

  • Thomas Mason

Student thesis: Phd


Substance misuse represents a global public health concern. The harms caused by substance use remain considerable internationally: 0.6 percent of the global adult population were estimated to have a substance use disorder in 2015, with 28 million disability-adjusted life years lost as a result of substance misuse and 17 million health years of life lost as a result of substance use disorders. However, there is relatively limited evidence on the economic impact of substance misuse on health care for England. An important pillar of recent substance misuse policy focusing on abstinence was a pay-for-performance (P4P) scheme which linked the income of providers of addiction treatment services to treatment outcomes. There is limited evidence on the effects of this scheme on in-treatment outcomes and wider related health care use and costs. This thesis applies econometric methods to two large administrative datasets for England between 2009-10 and 2015-16 in four main chapters. The first explores how the costs of substance misuse related hospital admissions have changed over the period 2009-10 to 2015-16 using Hospital Episode Statistics (HES), and how different costing approaches yield different estimates. Costs are shown to be overestimated by top-down approaches that apply average costs to specific diseases, and that estimates are highly sensitive to the use of primary and/or secondary diagnosis codes of substance misuse. The second uses an event study design to examine how hospital activity and related costs change around the time of the first occurrence. Annual costs per person are on average £2,953.65 higher in the event year than six years before, and remain £687.02 higher five years after. The third uses a difference-in-differences design to evaluate the impact of the introduction of P4P for providers of addiction treatment services on treatment outcomes using data from the National Drug Treatment Monitoring System (NDTMS) for 2010-11 to 2012-13. P4P reduced the probability of individuals completing substance misuse treatment successfully by 1.3 percentage points and increased the probability of individuals declining to continue with treatment by 0.9 percentage points. The final chapter considers the schemes impact on hospital admissions and costs, finding that individuals hospital admissions were 1.507 times higher in P4P areas after its introduction compared with non-intervention areas. The number and cost of substance misuse related hospital admissions increased substantially between 2009-10 and 2015-16. Furthermore, individuals hospital activity and costs increases prior to an admission diagnosed as being related to substance misuse - and is/are persistently higher after the event. Introduction of P4P for providers of addiction treatment services had negative effects on outcomes in the addiction treatment population and led to increased hospital admissions in P4P areas for a wider population with a record of substance misuse.
Date of Award31 Dec 2019
Original languageEnglish
Awarding Institution
  • The University of Manchester
SupervisorAndrew Jones (Supervisor), William Whittaker (Supervisor) & Matthew Sutton (Supervisor)


  • econometrics
  • illicit drugs
  • economics
  • policy evaluation
  • hospital costs
  • hospital admissions
  • substance misuse
  • P4P
  • pay-for-performance
  • financial incentives

Cite this