Linking inpatient clinical registry data to Medicare claims data using indirect identifiers

Duke Clinical Research Institute, Duke University School of Medicine, Durham, NC 27715, USA.
American heart journal (Impact Factor: 4.56). 07/2009; 157(6):995-1000. DOI: 10.1016/j.ahj.2009.04.002
Source: PubMed

ABSTRACT Inpatient clinical registries generally have limited ability to provide a longitudinal perspective on care beyond the acute episode. We present a method to link hospitalization records from registries with Medicare inpatient claims data, without using direct identifiers, to create a unique data source that pairs rich clinical data with long-term outcome data.
The method takes advantage of the hospital clustering observed in each database by demonstrating that different combinations of indirect identifiers within hospitals yield a large proportion of unique patient records. This high level of uniqueness also allows linking without advance knowledge of the Medicare provider number of each registry hospital. We applied this method to 2 inpatient databases and were able to identify 81% of 39,178 records in a large clinical registry of patients with heart failure and 91% of 6,581 heart failure records from a hospital inpatient database. The quality of the link is high, and reasons for incomplete linkage are explored. Finally, we discuss the unique opportunities afforded by combining claims and clinical data for specific analyses.
In the absence of direct identifiers, it is possible to create a high-quality link between inpatient clinical registry data and Medicare claims data. The method will allow researchers to use existing data to create a linked claims-clinical database that capitalizes on the strengths of both types of data sources.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Characteristics of stroke cases, acute stroke care, and outcomes after stroke differ according to geographical and cultural background. To provide epidemiological and clinical data on stroke care in South Korea, we analyzed a prospective multicenter clinical stroke registry, the Clinical Research Center for Stroke-Fifth Division (CRCS-5). Patients were 58% male with a mean age of 67.2±12.9 years and median National Institutes of Health Stroke Scale score of 3 [1-8] points. Over the 6 years of operation, temporal trends were documented including increasing utilization of recanalization treatment with shorter onset-to-arrival delay and decremental length of stay. Acute recanalization treatment was performed in 12.7% of cases with endovascular treatment utilized in 36%, but the proportion of endovascular recanalization varied across centers. Door-to-IV alteplase delay had a median of 45 [33-68] min. The rate of symptomatic hemorrhagic transformation (HT) was 7%, and that of any HT was 27% among recanalization-treated cases. Early neurological deterioration occurred in 15% of cases and were associated with longer length of stay and poorer 3-month outcomes. The proportion of mRS scores of 0-1 was 42% on discharge, 50% at 3 months, and 55% at 1 year after the index stroke. Recurrent stroke up to 1 year occurred in 4.5% of patients; the rate was higher among older individuals and those with neurologically severe deficits. The above findings will be compared with other Asian and US registry data in this article.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Studies on outcomes among patients with heart failure (HF) with preserved left ventricular ejection fraction (HFpEF), borderline left ventricular ejection fraction (HFbEF), and reduced left ventricular ejection fraction (HFrEF) remain limited. We sought to characterize mortality and readmission in patients with HF in the contemporary era. Get With The Guidelines-HF was linked to Medicare data for longitudinal follow-up. Patients were grouped into HFpEF (left ventricular ejection fraction [EF] ≥50%), HFbEF (40% ≤ EF < 50%), and HFrEF (EF <40%). Multivariable models were constructed to examine the relationship between EF and outcomes at 30 days and 1 year and to study trends over time. A total of 40,239 patients from 220 hospitals between 2005 and 2011 were included in the study: 18,897 (47%) had HFpEF, 5,626 (14%) had HFbEF, and 15,716 (39%) had HFrEF. In crude survival analysis, patients with HFrEF had slightly increased mortality compared with HFbEF and HFpEF. After risk adjustment, mortality at 1 year was not significantly different for HFrEF, HFbEF, and HFpEF (HFrEF vs HFpEF, hazard ratio [HR] 1.040 [95% CI 0.998-1.084], and HFbEF vs HFpEF, HR 0.967 [95% CI 0.917-1.020]). Patients with HFpEF had increased risk of all-cause readmission compared with HFrEF. Conversely, risk of cardiovascular and HF readmissions were higher in HFrEF and HFbEF compared with HFpEF. Among patients hospitalized with HF, patients with HFpEF and HFbEF had slightly lower mortality and higher all-cause readmission risk than patients with HFrEF, although the mortality differences did not persist after risk adjustment. Irrespective of EF, these patients experience substantial mortality and readmission highlighting the need for new therapeutic strategies. Copyright © 2014 Elsevier Inc. All rights reserved.
    American Heart Journal 11/2014; 168(5):721-730.e3. DOI:10.1016/j.ahj.2014.07.008 · 4.56 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Public health research is an integral part of the study of congenital heart disease. While this type of research has become more popular, particularly over the past decade, it has a history that stretches back to almost the beginnings of pediatric cardiology as a field. This review aims to introduce the concepts and methodologies of public health and how they relate to congenital heart disease, describe some of the challenges of traditional research methods in congenital heart disease, describe the history of public health research, and demonstrate the relevance of public health research, particularly databases, to pediatric cardiology fellows.
    Congenital Heart Disease 12/2014; 9(6). DOI:10.1111/chd.12235 · 1.20 Impact Factor

Full-text (2 Sources)

Available from
May 31, 2014