Development of dynamic health care delivery heatmaps for end-of-life cancer care: A cohort study

<p><b>Background:</b> Measures of variation in end-of-life treatment intensity across hospitals are typically summarized using unidimensional measures. These measures do not capture the full dimensionality of complex clinical care trajectories over time that are needed for actionable quality improvement efforts.</p>

<p><b>Objective:</b> To develop a novel longitudinal, multidimensional visual map of end-of-life care trajectories.</p>

<p><b>Methods:</b> We identified Medicare claims for fee-for-service beneficiaries with poor-prognosis cancers who died April- December 2016 and received the preponderance of treatment in the last 6-months of life at a National Cancer Institute or National Comprehensive Cancer Network (NCI-NCCN)-designated hospital. We ontologically modeled health care utilization in the last six months of life using systems theory into a heatmap image depicting system capabilities on the y-axis and time on the x-axis. We then use four measurement case studies to illustrate the value of longitudinal, multidimensional images over unidimensional metrics for guiding value-based, quality improvement initiatives.</p>

<p><b>Results:</b> We identified nine distinct patterns of end-of-life care from hospital-level dynamic utilization heatmaps based on signal intensity and patterns for inpatient, outpatient, and home-based hospice services. We illustrate that in most cases, multidimensional dynamic utilization heatmap patterns provide more information about care trajectories and highlight more heterogeneity than unidimensional measures.</p>

<p><b>Conclusions:</b> This study illustrates the feasibility of representing longitudinal and multidimensional end-of-life utilization dynamically as a heatmap. These heatmaps provide potentially actionable insights into hospital-level care delivery patterns, and the approach may generalize to other serious illness populations.</p>