Data-driven discovery of seasonally linked diseases from an Electronic Health Records system

Abstract

BackgroundPatterns of disease incidence can identify new risk factors for the disease or provide insight into the etiology. For example, allergies and infectious diseases have been shown to follow periodic temporal patterns due to seasonal changes in environmental or infectious agents. Previous work searching for seasonal or other temporal patterns in disease diagnosis rates has been limited both in the scope of the diseases examined and in the ability to distinguish unexpected seasonal patterns. Electronic Health Records (EHR) compile extensive longitudinal clinical information, constituting a unique source for discovery of trends in occurrence of disease. However, the data suffer from inherent biases that preclude a identification of temporal trends.MethodsMotivated by observation of the biases in this data source, we developed a method (Lomb-Scargle periodograms in detrended data, LSP-detrend) to find periodic patterns by adjusting the temporal information for broad trends in incidence, as well as seasonal changes in total hospitalizations. LSP-detrend can sensitively uncover periodic temporal patterns in the corrected data and identify the significance of the trend. We apply LSP-detrend to a compilation of records from 1.5 million patients encoded by ICD-9-CM (International Classification of Diseases, Ninth Revision, Clinical Modification), including 2,805 disorders with more than 500 occurrences across a 12 year period, recorded from 1.5 million patients.Results and conclusionsAlthough EHR data, and ICD-9 coded records in particular, were not created with the intention of aggregated use for research, these data can in fact be mined for periodic patterns in incidence of disease, if confounders are properly removed. Of all diagnoses, around 10% are identified as seasonal by LSP-detrend, including many known phenomena. We robustly reproduce previous findings, even for relatively rare diseases. For instance, Kawasaki disease, a rare childhood disease that has been associated with weather patterns, is detected as strongly linked with winter months. Among the novel results, we find a bi-annual increase in exacerbations of myasthenia gravis, a potentially life threatening complication of an autoimmune disease. We dissect the causes of this seasonal incidence and propose that factors predisposing patients to this event vary through the year.

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

“…While summary review of previous single disease studies may suggest that utilization rates tend to peak in the winter, our analysis of temporal variation in a diverse collection of clinical conditions has shown that summer peaks predominate. Furthermore, previous work has provided only a narrow understanding of temporal variation in hospital utilization whereas the largest study we identified was limited to only the top one-fifth of conditions with the largest mean monthly admission rate [15]. Here, we have dramatically expanded understanding of temporal variation through the study of 246 condition-specific hospital utilization rates, which together encompass nearly all medical conditions.…”

“…Such traditional approaches are unlikely to uncover unexpected temporal patterns that may be revealed only by broad data mining techniques applied across many conditions. While one previous study utilized a data mining approach, it was limited to the analysis of only one-fifth of diseases with the highest admission rates at a single institution, leaving the majority of diseases across a broader population unstudied [15]. In addition, 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.…”

The seasonality of diarrheal pathogens: A retrospective study of seven sites over three years

“…Seasonality was identified by estimating the power spectral density contained in the estimated incidence time-series data for each pathogen separately. Specifically, we utilize the Lomb-Scargle periodogram [24–26] to identify the spectral power for each resolvable frequency, which is constrained by the temporal length of GEMS and our two-week sampling interval. The Lomb-Scargle periodogram is especially useful for analyzing constrained time-series; the technique quantifies the uncertainty of detecting a specific periodicity [27] even with unevenly sampled data [25].…”

Evaluation of the secondary use of electronic health records to detect seasonal, holiday-related, and rare events related to traumatic injury and poisoning

“…While syndromic surveillance typically focuses on the detection and prevalence estimation of specific conditions, electronic health record databases can act as a generalized population health surveillance system, giving insight into previously unmonitored diseases. For instance, Melamed et al showed the utility of EHRs to link diseases to seasonal trends [16]. Other seasonal detection methods using EHR data have been used to model seasonal influenza outbreaks, seasonal blood pressure controls, and seasonal effects on early child development [17][18][19].…”