Development and validation of algorithms to predict left ventricular ejection fraction class from healthcare claims data.

AIMS: The use of large medical or healthcare claims databases is very useful for population-based studies on the burden of heart failure (HF). Clinical characteristics and management of HF patients differ according to categories of left ventricular ejection fraction (LVEF), but this information is often missing in such databases. We aimed to develop and validate algorithms to identify LVEF in healthcare databases where the information is lacking. METHODS AND RESULTS: Algorithms were built by machine learning with a random forest approach. Algorithms were trained and reinforced using the French national claims database [Système National des Données de Santé (SNDS)] and a French HF registry. Variables were age, gender, and comorbidities, which could be identified by medico-administrative code-based proxies, Anatomical Therapeutic Chemical codes for drug delivery, International Classification of Diseases (Tenth Revision) coding for hospitalizations, and administrative codes for any other type of reimbursed care. The algorithms were validated by cross-validation and against a subset of the SNDS that includes LVEF information. The areas under the receiver operating characteristic curve were 0.84 for the algorithm identifying LVEF ≤ 40% and 0.79 for the algorithms identifying LVEF < 50% and ≥50%. For LVEF ≤ 40%, the reinforced algorithm identified 50% of patients in the validation dataset with a positive predictive value of 0.88 and a specificity of 0.96. The most important predictive variables were delivery of HF medication, sex, age, hospitalization, and testing for natriuretic peptides with different orders of positive or negative importance according to the LVEF category. CONCLUSIONS: The algorithms identify reduced or preserved LVEF in HF patients within a nationwide healthcare claims database with high positive predictive value and low rates of false positives.

Feasibility and accuracy of linking a heart failure registry to the national claims database using indirect identifiers.

BACKGROUND: Heart failure (HF) registries include rich data on patient inclusion characteristics, but follow-up information is often incomplete. Medicoadministrative databases may provide less clinical information than registries, e.g. on left ventricular ejection fraction (LVEF), but long-term data are exhaustive and reliable. The combination of the two types of database is therefore appealing, but the feasibility and accuracy of such linking are largely unexplored. AIMS: To assess the feasibility and accuracy of linking an HF registry (FRESH; FREnch Survey on Heart Failure) with the French National Healthcare System database (SNDS). METHODS: A probabilistic algorithm was developed to link and match patient data included in the FRESH HF registry with anonymized records from the SNDS, which include: hospitalizations and diagnostic codes; all care-related reimbursements by national health system; and deaths. Consistency was assessed between deaths recorded in the registry and in the SNDS. A comparison between the two databases was carried out on several identifiable clinical characteristics (history of HF hospitalization, diabetes, atrial fibrillation, chronic bronchopneumopathy, severe renal failure and stroke) and on events during 1-year follow-up after inclusion. RESULTS: Of 2719 patients included in the FRESH registry (1049 during decompensation; 1670 during outpatient follow-up), 1885 could be matched with a high accuracy of 94.3% for deaths. Mortality curves were superimposable, including curves according to type of HF and LVEF. The rates of missing data in the FRESH registry were 2.3-8.4% for clinical characteristics and 17.5% for hospitalizations during follow-up. The discrepancy rate for clinical characteristics was 3-13%. Hospitalization rates were significantly higher in the SNDS than in the registry cohort. CONCLUSIONS: The anonymous matching of an HF research cohort with a national health database is feasible, with a significant proportion of patients being accurately matched, and facilitates combination of clinical data and a reduced rate of losses to follow-up.