Multi-state modelling of repeated hospitalisation and death in patients with heart failure: The use of large administrative databases in clinical epidemiology

Abstract

In chronic diseases like Heart Failure (HF), the disease course and associated clinical event histories for the patient population vary widely. To improve understanding of the prognosis of patients and enable health-care providers to assess and manage resources, we wish to jointly model disease progression, mortality and their relation with patient characteristics. We show how episodes of hospitalisation for disease-related events, obtained from administrative data, can be used as a surrogate for disease status. We propose flexible multi-state models for serial hospital admissions and death in HF patients, that are able to accommodate important features of disease progression, such as multiple ordered events and competing risks. Fully-parametric and semiparametric semi-Markov models are implemented using freely available software in R. The models were applied to a dataset from the administrative data bank of the Lombardia region in Northern Italy, which included 15,298 patients who had a first hospitalisation ending in 2006 and 4 years of follow up thereafter. This provided estimates of the associations of age and gender with rates of hospital admission and length of stay in hospital, and estimates of the expected total time spent in hospital over five years. For example, older patients and men were readmitted more frequently, though the total time in hospital was roughly constant with age. We also discuss the relative merits of parametric and semi-parametric multi-state models, and model assessment and comparison.

Effectiveness and cost-effectiveness of serum B-type natriuretic peptide testing and monitoring in patients with heart failure in primary and secondary care: an evidence synthesis, cohort study and cost-effectiveness model

“…through NYHA class) and making the probability of hospitalisation and death dependent on dysfunction would provide a more realistic representation of disease progression. For example, Ieva et al 106 have used routine administrative data to model the decrease in time to readmission with each successive admission and the association between age, sex and readmission. The simplicity of our model might lead to poor estimates of cost-effectiveness if BNP-guided therapy has a large effect (positive or negative) on functional decline among survivors.…”

Regional variation in hospitalisation and mortality in heart failure: comparison of England and Lombardy using multistate modelling

“…Other modelling options include hurdle models and resource buckets, which can be useful for some questions [11] but are not flexible enough to show progression of the underlying chronic disease. Standard Cox and marginal models are inappropriate for the same reason as they ignore the serial nature of admissions [12]; extensions of these can deal with the clustering of patients within hospitals and multiple events per patient but not the serial nature of those events, or they can account for the competing risk of death under certain assumptions but not the repeat events.…”

“…A second is that they employ the Markov assumption that future states depend only on the current state and time, but not on the whole history. This makes the method well suited to chronic disease progression modelling [12,13]. How likely the patient is to change state at any time t is described by the transition intensity, which may depend on time t and on a set of individual and time-dependent variables.…”

“…Compared with regression or survival analysis, these models are less familiar as they are not generally taught on statistics Master's courses and because implementation in standard software is either not possible or not as straightforward as logistic regression, for instance. To date there have been only a few studies using multistate models in heart failure, each limited to a small sample [14] or one country [12,15]. For context, we also use publicly available population estimates to calculate admission rates and provide estimated prevalences.…”

Dynamic clustering of hazard functions: an application to disease progression in chronic heart failure

“…The choice depends on the final aim of the analysis. In [11] an example of a multistate modelling strategy for the joint analysis of outcomes and hospital admissions in CHF patients is proposed. In that case the aim was to show a flexible approach, able to capture important features of admission-discharge dynamics such as multiple ordered events and the competing risks of death and hospitalisation, in a novel application based on data arising from the administrative database of Regione Lombardia.…”