Evaluating the effect of healthcare providers on the clinical path of heart failure patients through a semi-Markov, multi-state model

Modeling the multi-state natural history of rare diseases with heterogeneous individual patient data: A simulation study

Multi-state survival models are used to represent the natural history of a disease, forming the basis of a health technology assessment comparing a novel treatment to current practice. Constructing such models for rare diseases is problematic, since evidence sources are typically much sparser and more heterogeneous. This simulation study investigated different one-stage and two-stage approaches to meta-analyzing individual patient data (IPD) in a multi-state survival setting when the number and size of studies being meta-analyzed are small. The objective was to assess methods of different complexity to see when they are accurate, when they are inaccurate and when they struggle to converge due to the sparsity of data. Biologically plausible multi-state IPD were simulated from study- and transition-specific hazard functions. One-stage frailty and two-stage stratified models were estimated, and compared to a base case model that did not account for study heterogeneity. Convergence and the bias/coverage of population-level transition probabilities to, and lengths of stay in, each state were used to assess model performance. A real-world application to Duchenne Muscular Dystrophy, a neuromuscular rare disease, was conducted, and a software demonstration is provided. Models not accounting for study heterogeneity were consistently out-performed by two-stage models. Frailty models struggled to converge, particularly in scenarios of low heterogeneity, and predictions from models that did converge were also subject to bias. Stratified models may be better suited to meta-analyzing disparate sources of IPD in rare disease natural history/economic modeling, as they converge more consistently and produce less biased predictions of lengths of stay.

Lifespan and medical expenditure prognosis for cancer metastasis - a simulation modeling using semi-Markov process

BACKGROUND AND OBJECTIVE

A key reason of high mortality of cancers is attributed to the metastasized cancer, whereas, the medical expense for the treatment of cancer metastases generates heavily financial burden. The population size of metastases cases is small and comprehensive inferencing and prognosis is hard to conduct.

METHODS

Because metastases and finance state can develop dynamic transitions over time, this study proposes a semi-Markov model to perform risk and economic evaluation associated to major cancer metastasis (i.e., lung, brain, liver and lymphoma cancer) against rare cases. A nationwide medical database in Taiwan was employed to derive a baseline study population and costs data. The time until development of metastasis and survivability from metastasis, as well as the medical costs were estimated through a semi-Markov based Monte Carlo simulation.

RESULTS

In terms of the survivability and risk associated to metastatic cancer patients, 80% lung and liver cancer cases are tended to metastasize to other part of the body. The highest cost is generated by brain cancer-liver metastasis patients. The survivors group generated approximately 5 times more costs, in average, than the non-survivors group.

CONCLUSIONS

The proposed model provides a healthcare decision-support tool to evaluate the survivability and expenditure of major cancer metastases.

Dynamic risk stratification using Markov chain modelling in patients with chronic heart failure

AIMS

Risk changes with the progression of disease and the impact of treatment. We developed a dynamic risk stratification Markov chain model using artificial intelligence in patients with chronic heart failure (CHF).

METHODS AND RESULTS

We described the pattern of behaviour among 7496 consecutive patients assessed for suspected HF. The following mutually exclusive health states were defined and assessed every 4 months: death, hospitalization, outpatient visit, no event, and leaving the service altogether (defined as no event at any point following assessment). The observed figures at the first transition (4 months) weres 427 (6%), 1559 (21%), 2254 (30%), 1414 (19%), and 1842 (25%), respectively. The probabilities derived from the first two transitions (i.e. from baseline to 4 months and from 4 to 8 months) were used to construct the model. An example of the model's prediction is that at cycle 4, the cumulative probability of death was 14%; leaving the system, 37%; being hospitalized between 12 and 16 months, 10%; having an outpatient visit, 8%; and having no event, 31%. The corresponding observed figures were 14%, 41%, 10%, 15%, and 21%, respectively. The model predicted that during the first 2 years, a patient had a probability of dying of 0.19, and the observed value was 0.18.

CONCLUSIONS

A model derived from the first 8 months of follow-up is strongly predictive of future events in a population of patients with chronic heart failure. The course of CHF is more linear than is commonly supposed, and thus more predictable.

Similarity of competing risks models with constant intensities in an application to clinical healthcare pathways involving prostate cancer surgery

The recent availability of routine medical data, especially in a university-clinical context, may enable the discovery of typical healthcare pathways, that is, typical temporal sequences of clinical interventions or hospital readmissions. However, such pathways are heterogeneous in a large provider such as a university hospital, and it is important to identify similar care pathways that can still be considered typical pathways. We understand the pathway as a temporal process with possible transitions from a single initial treatment state to hospital readmission of different types, which constitutes a competing risks setting. In this article, we propose a multi-state model-based approach to uncover pathway similarity between two groups of individuals. We describe a new bootstrap procedure for testing the similarity of constant transition intensities from two competing risk models. In a large simulation study, we investigate the performance of our similarity approach with respect to different sample sizes and different similarity thresholds. The studies are motivated by an application from urological clinical routine and we show how the results can be transferred to the application example.