Discovery of temporal and disease association patterns in condition-specific hospital utilization rates.

Julian S Haimovich,Shu-Xia Li,Abbas Shojaee,Andreas Coppi,Harlan M Krumholz,Arjun K Venkatesh,Frederick Warner

doi:10.1371/journal.pone.0172049

Abstract

Identifying temporal variation in hospitalization rates may provide insights about disease patterns and thereby inform research, policy, and clinical care. However, the majority of medical conditions have not been studied for their potential seasonal variation. The objective of this study was to apply a data-driven approach to characterize temporal variation in condition-specific hospitalizations. Using a dataset of 34 million inpatient discharges gathered from hospitals in New York State from 2008–2011, we grouped all discharges into 263 clinical conditions based on the principal discharge diagnosis using Clinical Classification Software in order to mitigate the limitation that administrative claims data reflect clinical conditions to varying specificity. After applying Seasonal-Trend Decomposition by LOESS, we estimated the periodicity of the seasonal component using spectral analysis and applied harmonic regression to calculate the amplitude and phase of the condition’s seasonal utilization pattern. We also introduced four new indices of temporal variation: mean oscillation width, seasonal coefficient, trend coefficient, and linearity of the trend. Finally, K-means clustering was used to group conditions across these four indices to identify common temporal variation patterns. Of all 263 clinical conditions considered, 164 demonstrated statistically significant seasonality. Notably, we identified conditions for which seasonal variation has not been previously described such as ovarian cancer, tuberculosis, and schizophrenia. Clustering analysis yielded three distinct groups of conditions based on multiple measures of seasonal variation. Our study was limited to New York State and results may not directly apply to other regions with distinct climates and health burden. A substantial proportion of medical conditions, larger than previously described, exhibit seasonal variation in hospital utilization. Moreover, the application of clustering tools yields groups of clinically heterogeneous conditions with similar seasonal phenotypes. Further investigation is necessary to uncover common etiologies underlying these shared seasonal phenotypes.

Highlights

Epidemiological studies have identified temporal variation in hospitalization rates across different conditions, with respect to seasons
To characterize the temporal patterns embedded in the data, we developed an algorithm that combines the extraction of traditional indices of amplitude, period, and phase, with four newer indices of temporal variation: seasonal coefficient, mean oscillation width, trend coefficient, and trend linearity
Our analysis of condition-specific hospital utilization rates using a computational approach to characterize seasonal variation found that two-thirds of studied conditions demonstrate

Summary

Introduction

Epidemiological studies have identified temporal variation in hospitalization rates across different conditions, with respect to seasons. Significant seasonal temporal variation of hospital utilization rates has been shown for acute diseases such as stroke [1], venous thromboembolism [2, 3], influenza [4], acute myocardial infarction [5], as well as for chronic diseases including chronic obstructive pulmonary disease [6], heart failure [7, 8], atrial fibrillation [9], and psychiatric disorders [10] such as bipolar disorder and mood disorders [11] Each of these previously studied diseases exhibits annual utilization peaks in the winter. Previous work studying temporal patterns in hospital utilization rates has largely applied traditional statistical approaches that begin with a hypothesis regarding variation of a single disease driven by previously identified seasonal mediators These mediators range from infectious and microbiologic triggers [12] to environmental factors such as temperature [13] and pollution [14]. Past studies have reduced the complexity of temporal pattern analysis to three simple indices, namely amplitude, period, and phase, which together provide only a limited characterization of temporal variation

Objectives

Methods

Results

Conclusion