Tailoring integrated care services for high-risk patients with multiple chronic conditions: a risk stratification approach using cluster analysis

Abstract

Background: The purpose of this study was to produce a risk stratification within a population of high-risk patients with multiple chronic conditions who are currently treated under a case management program and to explore the existence of different risk subgroups. Different care strategies were then suggested for healthcare reform according to the characteristics of each subgroup. Methods: All high-risk multimorbid patients from a case management program in the Navarra region of Spain were included in the study (n = 885). A 1-year mortality risk score was estimated for each patient by logistic regression. The population was then divided into subgroups according to the patients' estimated risk scores. We used cluster analysis to produce the stratification with Ward's linkage hierarchical algorithm. The characteristics of the resulting subgroups were analyzed, and post hoc pairwise tests were performed. Results: Three distinct risk strata were found, containing 45, 38 and 17% of patients. Age increased from cluster to cluster, and functional status, clinical severity, nursing needs and nutritional values deteriorated. Patients in cluster 1 had lower renal deterioration values, and patients in cluster 3 had higher rates of pressure skin ulcers, higher rates of cerebrovascular disease and dementia, and lower prevalence rates of chronic obstructive pulmonary disease. Conclusions: This study demonstrates the existence of distinct subgroups within a population of high-risk patients with multiple chronic conditions. Current case management integrated care programs use a uniform treatment strategy for patients who have diverse needs. Alternative treatment strategies should be considered to fit the needs of each patient subgroup.

Profile of multimorbidity in outpatients attending public healthcare settings

“…Further, the findings offer greater insights into future public health research, particularly towards vulnerability stratification within multimorbidity for patient management. [ 36 ]…”

Use of Latent Class Analysis and k-Means Clustering to Identify Complex Patient Profiles

“…Other large studies in this area have been limited by reliance on administrative data for large general populations (leading to self-evident divisions between healthier and less healthy categories), 30 , 31 , 32 and studies with greater depth of clinical data have been limited to smaller populations. 33 , 34 , 35 Moreover, studies relying largely on diagnosis codes, while able to identify which medical conditions cooccur, may lack the ability to distinguish patients based on severity of illness, functional status, medication needs, or nonmedical factors that are associated with health outcomes and therefore are limited in their ability to suggest specific care innovations. 36 , 37 , 38 , 39 Results from diagnosis code analyses are useful for identifying which conditions are enriched or cooccur in high-cost conditions (eg, end-stage renal disease, diabetes with multiple comorbidities, and acute on chronic illness) but are less helpful for designing care around patient-centered phenotypes (eg, frail elderly adults, chronic pain management).…”

Exploring and Identifying Prognostic Phenotypes of Patients with Heart Failure Guided by Explainable Machine Learning

“…Ward’s method was used as a linkage criterion, which minimized the sum of squared differences within all clusters [ 21 ]. The optimal number of clusters was determined based on dendrogram [ 15 ], scree plot [ 22 ], and elbow methods [ 23 ]. Dimension reduction was performed using uniform manifold approximation and projection [ 24 ].…”

“…Segar et al used a penalized finite-mixture-model-based clustering analysis to successfully identify three phenotypes with distinct clinical characteristics and long-term outcomes of patients with HF with preserved ejection fraction [ 14 ]. Bretos-Azcona et al demonstrated the existence of distinct three risk subgroups within population of high-risk multiple chronic condition patients using a clustering method, and suggested subgroup-specific treatment strategy instead of a uniform one [ 15 ]. In addition, Stevens et al declared unsupervised ML could provide a basis for homogenization and was expected to guide personalized intervention [ 16 ].…”