Structured, Harmonized, and Interoperable Integration of Clinical Routine Data to Compute Heart Failure Risk Scores

Abstract

Risk prediction in patients with heart failure (HF) is essential to improve the tailoring of preventive, diagnostic, and therapeutic strategies for the individual patient, and effectively use health care resources. Risk scores derived from controlled clinical studies can be used to calculate the risk of mortality and HF hospitalizations. However, these scores are poorly implemented into routine care, predominantly because their calculation requires considerable efforts in practice and necessary data often are not available in an interoperable format. In this work, we demonstrate the feasibility of a multi-site solution to derive and calculate two exemplary HF scores from clinical routine data (MAGGIC score with six continuous and eight categorical variables; Barcelona Bio-HF score with five continuous and six categorical variables). Within HiGHmed, a German Medical Informatics Initiative consortium, we implemented an interoperable solution, collecting a harmonized HF-phenotypic core data set (CDS) within the openEHR framework. Our approach minimizes the need for manual data entry by automatically retrieving data from primary systems. We show, across five participating medical centers, that the implemented structures to execute dedicated data queries, followed by harmonized data processing and score calculation, work well in practice. In summary, we demonstrated the feasibility of clinical routine data usage across multiple partner sites to compute HF risk scores. This solution can be extended to a large spectrum of applications in clinical care.

Keywords: HiGHmed; clinical routine data; heart failure; medical data integration center; medical informatics initiative; openEHR; risk prediction scores; semantic interoperability.

Figure 1

Number of available data items for HF score calculation at each of the five participating partner sites. Entities over five openEHR template classes are shown (see color code) and the number of successfully computed HF scores. The upper boundary (i.e., the number of considered individual patients) is shown as dashed line.

Figure 2

The split violin plots for the MAGGIC score and the BCN BioHFv1 score (1- and 3-year mortality risk, respectively) across the different clinical sites. Black dots show median values and thin black lines show upper and lower quartiles. We either stratified by (a) patient status (inpatient vs. outpatient) or by (b) patient sex (female vs. male). Partner site codes: B = Berlin, G = Göttingen, H = Hannover, HD = Heidelberg, WU = Würzburg.

Figure 3

Conditional inference tree for MAGGIC score features (see Table 1). Score features are used to assign patients to partner sites and to pinpoint differences in the site-specific patient cohorts. Terminal nodes show patient proportions over sites and the total number of patients, respectively.

Figure 4

Conditional inference tree for BioHFv1 score features (see Table 1). Score features are used to assign patients to partner sites and to pinpoint differences in the site-specific patient cohorts. Terminal nodes show patient proportions over sites and the total number of patients, respectively.