Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.789310
Title: Using whole disease modelling to inform resource allocation decisions in schizophrenia services
Author: Li, Huajie
ISNI:       0000 0004 8500 6443
Awarding Body: King's College London
Current Institution: King's College London (University of London)
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Restricted access.
Access from Institution:
Abstract:
This thesis explores the need for, feasibility and value of building a whole disease model (WDM) of schizophrenia and its management in the UK. Conventional health economic models are 'piecewise' in that they typically address a single decision problem at a specific decision point in a care pathway. This can lead to inconsistencies between decisions and such economic analyses may not achieve their intended objective - the maximisation of health given a fixed budget constraint. An alternative system-level modelling approach - whole disease modelling - involves developing a health economic model of whole disease and treatment pathways which support the economic analysis of technologies and services for prevention, diagnosis, treatment and follow-up within a single consistent mathematical model. System-level modelling approaches (including WDMs) have been applied to several disease areas, including cancer, metabolic diseases and cardiovascular diseases. However, they have not been applied to mental health disorders, such as schizophrenia. A systematic literature review of existing model-based economic analyses for schizophrenia services revealed that: (1) there are no existing WDMs for schizophrenia; (2) there are few model-based economic analyses for non-pharmacological interventions; (3) existing models are generally poor quality; (4) existing models cannot support the joint assessment of options for service investment and disinvestment. To address these problems, this thesis focuses on the design, development and evaluation of a novel WDM of schizophrenia services in the UK. The feasibility and value of the schizophrenia WDM in informing resource allocation decisions was evaluated by using the model to assess multiple topics considered within existing NICE guidelines on the management of schizophrenia. The schizophrenia WDM was implemented as a patient-level discrete event simulation model. The choice of modelling technique was informed by a literature review of modelling techniques adopted by previous system-level models and the application of published model selection tools. Conceptual models underpinning the WDM were developed using a methodological framework specially adapted for this thesis. Evidence to inform the model parameters was obtained from the best available sources. Extensive validation and verification activities were undertaken to ensure that the model was robust. The WDM was used to conduct cost-effectiveness analyses for five NICE guideline topics (Topic A, B, C, D and E), covering five different components of the schizophrenia care pathway and involving 19 different interventions. Outcomes were measured in terms of quality-adjusted life years (QALYs) and costs included those relevant from the health and social care perspective. Two decision rules were adopted to assess the cost-effectiveness of alternative options: (1) a piecewise cost per QALY threshold rule (Decision Rule 1), and (2) a disease-level constrained maximisation of QALYs decision rule, which is subject to a fixed budget constraint (Decision Rule 2). The WDM results suggest that the current schizophrenia service configuration is not optimal. Cost savings and additional QALYs can be gained by: (1) replacing current interventions with more cost-effective interventions; (2) increasing the level of implementation of currently recommended interventions. Results using Decision Rule 1 suggest that the most cost-effective interventions for each topic are: cognitive behavioural therapy (CBT) for people at clinical high risk of psychosis (Topic A); a mix of hospital admission and crisis resolution and home treatment team (CRHT) for people with acute episode of psychosis (Topic B); amisulpride, risperidone or olanzapine as the first-line oral antipsychotic medication for people with first-episode psychosis (FEP) (Topic C); family intervention combined with antipsychotic medication for people with FEP (Topic D); and clozapine for people with treatment resistant schizophrenia (Topic E). The use of Decision Rule 2 leads to a different conclusion for Topic C, with quetiapine becoming the most cost-effective option. The results also suggest that the following options for improving adherence to the NICE schizophrenia guideline recommendations are likely to result in cost savings and QALY gains compared with current practice: (1) increasing the provision of CBT from current levels of 41.01% to 100% for people at clinical high risk of psychosis; (2) increasing the provision of family intervention from current levels of 30.98% to 100% for people with FEP; and (3) providing clozapine to people with treatment-resistant schizophrenia with no delay compared to current provision with a delay of 3.98 years. This thesis demonstrates that it is feasible to develop a WDM for schizophrenia, although a significant initial investment of time and resources was required; overall, the schizophrenia WDM took approximately 12 months to develop (full-time equivalent) and 4 months to run. Compared with conventional piecewise models, WDMs offer several benefits, such as addressing more decision problems using a single consistent model, flexibility around the choice of decision rules, and the inclusion of interactions between interventions across the pathway. The schizophrenia WDM shows that the use of alternative decision rules and changing the current service configuration can alter the cost-effectiveness conclusions for certain topics. Three priorities for future research are warranted: (1) improving the evidence base for schizophrenia models; (2) re-use of the WDM to evaluate other decision problems within schizophrenia services; and (3) exploring methods for reducing the development time of WDMs.
Supervisor: Byford, Sarah ; MacCabe, James Hunter Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.789310  DOI: Not available
Share: