Using gradient boosting with stability selection on health insurance claims data to identify disease trajectories in chronic obstructive pulmonary disease

Ploner, Tina; Heß, S.; Grum, Marcus; Drewe-Boß, Philipp; Walker, Jochen

doi:10.1177/0962280220938088

Cited by 8 publications

(4 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Researchers may be interested in moving beyond descriptive research when investigating trajectories of social and health outcomes and instead adopt a predictive lens if different states or trajectories are already defined by the topic, for example, disease severity or educational or occupational level. A study investigated predictors of chronic obstructive pulmonary disease of highest severity and common disease trajectories with gradient boosting and a shifting time window approach in health claims data: The authors identified a number of diagnoses (e.g., respiratory failure), medications (e.g., anticholinergic drugs), and procedures associated with a subsequent chronic obstructive pulmonary disease diagnosis of highest severity ( 91 ). The temporal patterns detected in this study rather represent order of health care–relevant diseases and should not be interpreted in the sense of causal pathways ( 91 ).…”

Section: For Predictionmentioning

confidence: 99%

“…A study investigated predictors of chronic obstructive pulmonary disease of highest severity and common disease trajectories with gradient boosting and a shifting time window approach in health claims data: The authors identified a number of diagnoses (e.g., respiratory failure), medications (e.g., anticholinergic drugs), and procedures associated with a subsequent chronic obstructive pulmonary disease diagnosis of highest severity ( 91 ). The temporal patterns detected in this study rather represent order of health care–relevant diseases and should not be interpreted in the sense of causal pathways ( 91 ). In other contexts, detected temporal patterns may be more robust.…”

Section: For Predictionmentioning

confidence: 99%

See 1 more Smart Citation

Mapping of machine learning approaches for description, prediction, and causal inference in the social and health sciences

et al. 2022

View full text Add to dashboard Cite

Machine learning (ML) methodology used in the social and health sciences needs to fit the intended research purposes of description, prediction, or causal inference. This paper provides a comprehensive, systematic meta-mapping of research questions in the social and health sciences to appropriate ML approaches by incorporating the necessary requirements to statistical analysis in these disciplines. We map the established classification into description, prediction, counterfactual prediction, and causal structural learning to common research goals, such as estimating prevalence of adverse social or health outcomes, predicting the risk of an event, and identifying risk factors or causes of adverse outcomes, and explain common ML performance metrics. Such mapping may help to fully exploit the benefits of ML while considering domain-specific aspects relevant to the social and health sciences and hopefully contribute to the acceleration of the uptake of ML applications to advance both basic and applied social and health sciences research.

show abstract

Section: For Predictionmentioning

confidence: 99%

Section: For Predictionmentioning

confidence: 99%

Mapping of machine learning approaches for description, prediction, and causal inference in the social and health sciences

et al. 2022

View full text Add to dashboard Cite

show abstract

“…From the learned parameters, a graph of disease-symptom connections was elicited, and the developed knowledge graphs were assessed and approved, with consent, against Google's Temporal patterns in patient disease trajectories are either disregarded or only taken into account by assessing the temporal directionality of identified co-morbidity pairs [14], [15], [16], Which concludes naturally patients undergo different symptoms at different time instances on disease trajectories (stages of the disease). In this experiment, the temporal pattern for each disease as the preliminary basis of a disease prediction model is utilized.…”

Section: A Data Collection and Preparationmentioning

confidence: 99%

Dynamic Spatial-Temporal Graph Model for Disease Prediction

Senthilkumar,

Gupte,

2022

IJACSA

View full text Add to dashboard Cite

Advances in the field of Neural Networks, especially Graph Neural Networks (GNNs) has helped in many fields, mainly in the areas of Chemistry and Biology where recognizing and utilising hidden patterns is of much importance. In Graph Neural Networks, the input graph structures are exploited by using the dependencies formed by the nodes. The data can also be transformed in the form of graphs which can then be used in such models. In this paper, a method is proposed to make appropriate transformations and then to use the structure to predict diseases. Current models in disease prediction do not fully use the temporal features that are associated with diseases, such as the order of the occurrence of symptoms and their significance. In the proposed work, the presented model takes into account the temporal features of a disease and represents it in terms of a graph to fully utilize the power of Graph Neural Networks and Spatial-Temporal models which take into consideration of the underlying structure that change over time. The model can be efficiently used to predict the most likely disease given a set of symptoms as input. The model exhibits the best algorithm based on its accuracy. The accuracy of the algorithm is determined by the performance on the given dataset. The proposed model is compared with the existing baseline models and proves to be outstanding and more promising in the disease prediction.

show abstract

“…For example, claims data are used to optimize health service provisions, 1 estimate the prevalence and incidence of diseases, 2 or identify patients at risk of hospital readmission. 3 While claims data have been extensively used for disease prediction, [4][5][6] much less attention has been paid to their vast amount of (implicit) information on disease dynamics over time, such as (changes in) medication, hospital stays, or the frequency of physician consultations (but see Ploner et al 7 ). In this contribution, we thus draw on these comprehensive datasets to model the temporal courses of diseases and to learn about patients' health conditions over time.…”

Section: Introductionmentioning

confidence: 99%

Markov‐modulated marked Poisson processes for modeling disease dynamics based on medical claims data

Mews

Surmann

Hasemann

et al. 2023

Statistics in Medicine

View full text Add to dashboard Cite

We explore Markov‐modulated marked Poisson processes (MMMPPs) as a natural framework for modeling patients' disease dynamics over time based on medical claims data. In claims data, observations do not only occur at random points in time but are also informative, that is, driven by unobserved disease levels, as poor health conditions usually lead to more frequent interactions with the health care system. Therefore, we model the observation process as a Markov‐modulated Poisson process, where the rate of health care interactions is governed by a continuous‐time Markov chain. Its states serve as proxies for the patients' latent disease levels and further determine the distribution of additional data collected at each observation time, the so‐called marks. Overall, MMMPPs jointly model observations and their informative time points by comprising two state‐dependent processes: the observation process (corresponding to the event times) and the mark process (corresponding to event‐specific information), which both depend on the underlying states. The approach is illustrated using claims data from patients diagnosed with chronic obstructive pulmonary disease by modeling their drug use and the interval lengths between consecutive physician consultations. The results indicate that MMMPPs are able to detect distinct patterns of health care utilization related to disease processes and reveal interindividual differences in the state‐switching dynamics.

show abstract

Using gradient boosting with stability selection on health insurance claims data to identify disease trajectories in chronic obstructive pulmonary disease

Cited by 8 publications

References 26 publications

Mapping of machine learning approaches for description, prediction, and causal inference in the social and health sciences

Mapping of machine learning approaches for description, prediction, and causal inference in the social and health sciences

Dynamic Spatial-Temporal Graph Model for Disease Prediction

Markov‐modulated marked Poisson processes for modeling disease dynamics based on medical claims data

Contact Info

Product

Resources

About