Post-stroke cognitive impairment: a systematic review of disease progression models

Abstract

Background: The aim was to review models of post-stroke cognitive impairment (CI) and dementia disease progression, to inform development of the StrokeCog epidemiological modelling platform for use in planning and economic decision-modelling for psychological services for stroke. Methods: Pubmed, EMBASE and the NHS Economic Evaluations Database were searched for relevant articles published in English since 2000. Included articles described an epidemiological or economic model tracking disease progression of CI and/or stroke over time, e.g., Markov or microsimulation models. Models that focussed on prevention rather than disease progression, or that focussed solely on resource utilisation, were excluded. Titles and abstracts were screened by one reviewer, followed by the full text of potentially relevant articles. Data was abstracted from eligible articles using a standard form. Models were evaluated against pre-specified criteria for the StrokeCog model, agreed by expert consensus. Findings: 4,638 abstracts were screened, followed by full text review of 345 articles. There were 61 eligible articles for CI and 47 for stroke (only two articles modelled both stroke and CI). Two modelling options were identified in most eligible articles: 1) a cohort-level state-transition model with disease states defined by categories of CI and/or disability, and 2) an individual-level discrete event simulation model with continuous modelling of CI, disability, behaviour and/or mood. In stroke models, transitions between disease states were assumed to be dependent on stroke recurrence. Discussion: A systematic, rigorous review of current modelling approaches was used to develop options for an epidemiological model of pathways of post-stroke CI over time.